As I drove home from work yesterday, I listened to the reports on Bloomberg Radio about a possible new company called MorganStanleyWachovia or MorganStanleyCiti.
Most people think about the financial ramifications of such a marriage. I think of the ramifications for the IT departments, especially focusing on trading technology.
I think that both of the combinations would be extremely difficult to integrate as far as IT goes. The IT culture of Morgan Stanley is so much different that any other place (with maybe the exception of Goldman). Morgan is very much of a build-it-yourself culture. They have their own messaging system (CPS) and their own ticker plant (Filter). The mad Russian scientists who populate the various IT departments pride themselves on the code that they right. It is much much different than the IT culture that you find in large banks like Citi, Wachovia, Well Fargo, etc.
It is public information that a lot of Morgan Stanley talent has recently migrated to Citi. Vikram Pandit is ex-Morgan, and you see many more ex-Morganites starting to populate the executive ranks. So, the "pump is primed" at Citi for a Morgan Stanley merger, at least at the executive levels. But the IT cultures are so radically different, both in terms of culture and technology, that I think that it will take a very long time to integrate.
Wachovia always had the relatively laid-back Charlotte way of doing things. Lots of legacy technology swimming around their halls. A few years ago, Wachovia established a base in New York, and was able to lure a lot of good Goldman people away. But, there is still that back-and-forth between New York and Charlotte, and most of the Wachovia people that I know in New York have to fly to Charlotte on a regular basis to get their marching directions. This is not the Morgan way of doing things.
BankOfAmericaMerrill is going to be another fun ride on the IT-integration-ferris wheel. Merrill has some really good technology. Some is very new and some is really old (the CICS-based systems that were written 20 years ago are still fairly important).
This is a time when Integration Architecture might be the hottest skill set on Wall Street. I expect boom times for companies like Accenture, IBM, Capco, and the other large consulting firms that specialize in technology integration.
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
Thursday, September 18, 2008
Wednesday, September 17, 2008
We are Hiring
I am going to indulge in what some of you might think is boorish and vulture-like behavior. My heart bleeds for the people at Lehman, Bear, and Merrill who have been displaced, either recently or in the near future. However, people have to live and they need to support their families. So, here I go ....
My company has openings in most of the departments within Equities .... market making, trading systems, risk, etc. We are interested in all skill sets, most notably Java and C++. We need great, hard-core developers. These are not management positions.
If you are interested, please send me a resume. I do not profit personally from this. I am just interested in seeing us with the best talent out there, and catching up to our remaining competitors.
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
My company has openings in most of the departments within Equities .... market making, trading systems, risk, etc. We are interested in all skill sets, most notably Java and C++. We need great, hard-core developers. These are not management positions.
If you are interested, please send me a resume. I do not profit personally from this. I am just interested in seeing us with the best talent out there, and catching up to our remaining competitors.
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
Review of the Gartner 2008 Conference on Event Processing
I spent an interesting two days in scenic Stamford, Connecticut on September 15 and 16, 2008. It was my second venture to a Gartner conference solely devoted to event processing, the first being in Orlando in 2007 when we were CEP neophytes. This time, I was not in learning mode .... the sessions that I attended were designed to give me an insight into what other financial companies are doing with CEP, and what the future direction of CEP might be.
The conference seemed a bit smaller than last year's conference, maybe because this year, it was not preceded directly by the Gartner conference on Business Intelligence. Some of the vendors were lamenting the lack of prospects. Indeed, the conference seemed much more heavily stacked with vendor participants than with potential customers. This speculation was confirmed when I saw a list of the attendees. I would estimate that, out of about 200 attendees, maybe 2o-25% were customers. I was one of 9 attendees from financial services firms, with most of the other attendees coming from fairly small companies. Does this mean that financial companies have already made their decisions concerning CEP, or are the financial companies being more prudent with their spending on conferences and travel? The next year should be a very interesting one for the vendors of CEP products, especially ones that are targeting financial firms.
Mary Knox started off with her usually overview of adopting CEP in financial services. A fairly good overview, with many valid points. I also notice that Mary has been reading certain blogs, as some of this information was included in her presentation. I questioned Mary's numbers on CEP spending, as it showed an extremely rosy outlook for the CEP marketplace, an outlook which may have been valid at last year's conference, but may no longer be valid in these times.
Mark Palmer of Streambase gave a great presentation on Smart Order Routing. I feared that it was going to be an informercial for Streambase, especially since, as I walked into the conference room, a Streambase rep thrust a USB flash drive at me which contained the latest incarnation of the Streambase platform. However, Mark presented a very good presentation without really harping on Streambase, and to tell you the truth, I like the direction where they are heading. It's pretty obvious that Mark is bringing a lot of the direction from Apama to Streambase in terms of developing financial verticals. I also hope that Mark will have a positive effect on the Streambase culture, and that their sales and marketing team will adopt a lower-key approach.
Mark's presentation was followed by Robert Almgren, who used to be with Bank of America before striking out on his own (???). Last year, Robert gave a fascinating intro to Algo trading. This year, he gave some in-depth looks at the architecture of an algo trading system, something that he was probably not allowed to do last year. Streambase is sponsoring some work that he and Bill Harts are doing in Smart Order Routing, and I am sure that the results will make their way back into the Streambase platform. Streambase's direction of totally embracing the financial services industry may or may not pay off in the future; from what a few of the vendors have told me, the interest in CEP in the hedge fund sector has cooled off somewhat.
Louis Morgan from HG Trading pretty much rehashed last year's presentation. HG Trading specializes in High Frequency Trading. Their average holding time for a stock is 7 minutes, and they never hold a position overnight. He mentioned that trading volume have gone up substantially, while the latency of his system has gone from 20ms to under 2ms. HG trading is a three person shop .... Lou, his developer, and a quant. They are extremely agile, and not afraid to spend money where it is needed. For instance, they spend money on co-location for their servers. They have 6 Apama correlators running. It was interesting that Lou gave his talk around the time the market opened on Monday (the morning of the Merrill and Lehman announcements), and he would have been totally forgiven if he has to cancel his presentation and attend to business.
Another good talk with given by Albert Doolittle of George Weiss, a small trading firm in Hartford, Connecticut. I saw many parallels between Albert's experience and my own .... he just jumped into CEP last year, did the trade show circuit, picked Aleri, did a POC, and now he has a CEP system up and running.
I think that my own talk went fairly well, and that I save the attendees some things to ponder.
It was nice to see some of the luminaries of the CEP world, including Opher, and to reconnect with many of the CEP vendors. Ron from Aleri introduced me to the Black Martini (Vodka, Blue Curacao, and Chamborde), and I know that he was moving a few steps slower the following morning.
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
The conference seemed a bit smaller than last year's conference, maybe because this year, it was not preceded directly by the Gartner conference on Business Intelligence. Some of the vendors were lamenting the lack of prospects. Indeed, the conference seemed much more heavily stacked with vendor participants than with potential customers. This speculation was confirmed when I saw a list of the attendees. I would estimate that, out of about 200 attendees, maybe 2o-25% were customers. I was one of 9 attendees from financial services firms, with most of the other attendees coming from fairly small companies. Does this mean that financial companies have already made their decisions concerning CEP, or are the financial companies being more prudent with their spending on conferences and travel? The next year should be a very interesting one for the vendors of CEP products, especially ones that are targeting financial firms.
Mary Knox started off with her usually overview of adopting CEP in financial services. A fairly good overview, with many valid points. I also notice that Mary has been reading certain blogs, as some of this information was included in her presentation. I questioned Mary's numbers on CEP spending, as it showed an extremely rosy outlook for the CEP marketplace, an outlook which may have been valid at last year's conference, but may no longer be valid in these times.
Mark Palmer of Streambase gave a great presentation on Smart Order Routing. I feared that it was going to be an informercial for Streambase, especially since, as I walked into the conference room, a Streambase rep thrust a USB flash drive at me which contained the latest incarnation of the Streambase platform. However, Mark presented a very good presentation without really harping on Streambase, and to tell you the truth, I like the direction where they are heading. It's pretty obvious that Mark is bringing a lot of the direction from Apama to Streambase in terms of developing financial verticals. I also hope that Mark will have a positive effect on the Streambase culture, and that their sales and marketing team will adopt a lower-key approach.
Mark's presentation was followed by Robert Almgren, who used to be with Bank of America before striking out on his own (???). Last year, Robert gave a fascinating intro to Algo trading. This year, he gave some in-depth looks at the architecture of an algo trading system, something that he was probably not allowed to do last year. Streambase is sponsoring some work that he and Bill Harts are doing in Smart Order Routing, and I am sure that the results will make their way back into the Streambase platform. Streambase's direction of totally embracing the financial services industry may or may not pay off in the future; from what a few of the vendors have told me, the interest in CEP in the hedge fund sector has cooled off somewhat.
Louis Morgan from HG Trading pretty much rehashed last year's presentation. HG Trading specializes in High Frequency Trading. Their average holding time for a stock is 7 minutes, and they never hold a position overnight. He mentioned that trading volume have gone up substantially, while the latency of his system has gone from 20ms to under 2ms. HG trading is a three person shop .... Lou, his developer, and a quant. They are extremely agile, and not afraid to spend money where it is needed. For instance, they spend money on co-location for their servers. They have 6 Apama correlators running. It was interesting that Lou gave his talk around the time the market opened on Monday (the morning of the Merrill and Lehman announcements), and he would have been totally forgiven if he has to cancel his presentation and attend to business.
Another good talk with given by Albert Doolittle of George Weiss, a small trading firm in Hartford, Connecticut. I saw many parallels between Albert's experience and my own .... he just jumped into CEP last year, did the trade show circuit, picked Aleri, did a POC, and now he has a CEP system up and running.
I think that my own talk went fairly well, and that I save the attendees some things to ponder.
It was nice to see some of the luminaries of the CEP world, including Opher, and to reconnect with many of the CEP vendors. Ron from Aleri introduced me to the Black Martini (Vodka, Blue Curacao, and Chamborde), and I know that he was moving a few steps slower the following morning.
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
Goodbye Lehman and Merrill
I could not sleep last Sunday night, the night before the Gartner CEP conference was to start. I was transfixed to CNBC for the entire night. And, I was really depressed ....
Why? Merrill Lynch was disappearing as a firm. One of the most storied financial firms in the world, and a widely recognized brand name.
My father was a broker for Merrill Lynch for 25 years, first starting in a little branch office on Austin Street in Forest Hills, Queens. One of his friends at the office, a guy who used to come over our house and play Pinochle with my dad, was a guy named Dave Kamamsky. Big Dave eventually rose through the ranks at Merrill to become the CEO of Merrill during the go-go years of the 1990's.
My wife worked for Merrill Lynch for about 10 years, starting off as a CICS/Cobol developer and rising to run several of the Equity trading systems there.
Seeing Merrill being swallowed up by Bank of America is like losing an arm. In a year, we have lost three of the major financial institutions in New York. Who knows how long Morgan Stanley will be around ... as I write this, I am watching Charles Gasparino on CNBC saying that John Mack is weighing whether or not to keep Morgan Stanley independent.
These are life-changing times we are in ....
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
Why? Merrill Lynch was disappearing as a firm. One of the most storied financial firms in the world, and a widely recognized brand name.
My father was a broker for Merrill Lynch for 25 years, first starting in a little branch office on Austin Street in Forest Hills, Queens. One of his friends at the office, a guy who used to come over our house and play Pinochle with my dad, was a guy named Dave Kamamsky. Big Dave eventually rose through the ranks at Merrill to become the CEO of Merrill during the go-go years of the 1990's.
My wife worked for Merrill Lynch for about 10 years, starting off as a CICS/Cobol developer and rising to run several of the Equity trading systems there.
Seeing Merrill being swallowed up by Bank of America is like losing an arm. In a year, we have lost three of the major financial institutions in New York. Who knows how long Morgan Stanley will be around ... as I write this, I am watching Charles Gasparino on CNBC saying that John Mack is weighing whether or not to keep Morgan Stanley independent.
These are life-changing times we are in ....
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
Saturday, September 13, 2008
Optimizing Performance in a CEP System
Here are some lessons learned that helped us reduce message backup from our Tibco EMS broker into our CEP Input Server.
1) Make sure that you have fast networking between the Tibco broker and your servers.
We just happened to find out that the hardware folks gave us legacy 100MB connections between our Tibco EMS broker and our CEP Input Server. We asked them to upgrade us to GigE immediately. Unless you have personally installed your hardware, never assume anything in your infrastructure.
2) It might be helpful to turn the "AutoAcknowledge" flag off in your Tibco session.
To do this, change
this.m_connection.CreateSession(false, SessionMode.AutoAcknowledge);
to
this.m_connection.CreateSession(false, SessionMode.NoAcknowledge);
Be careful about using No Acks. What is the impact to your CEP system if you happen to drop a message?
3) Implement threaded queues on the input from Tibco to your input server, and on the output from your input server to the CEP system.
If you need more performance, change the single threaded queue that reads messages from Tibco to a load-balanced multiple queues, each with its own thread. The load balancer can partition by Tibco topic, by ticker, or by OrderId.
Of course, you can always use the built-in adapters from your CEP vendor. But that may come at an extra monetary cost, so check with your CEP vendor to see if a certain input adapter comes free of charge. Also, the more of the CEP vendor's infrastructure you use, the more it ties you into the CEP vendor.
4) Try to do some pre-filtering of messages in your input server before passing them on to the CEP engine.
Recently, we were able to reduce the CPU usage of Coral8 by about 50% (according to our Coral8 guy) by filtering out FIX messages that we knew that Coral8 was not interested in.
5) Continually optimize your code. Just this morning, I did a code review of the input server, and I was able to suggest a change which reduced one hashtable lookup on each FIX message. I know that there are tons of other places in our code that can do with some optimization. If you need to parse FIX messages into C# objects, then do some investigation and benchmarking of FIX engines.
6) Use the Server Garbage collector
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
1) Make sure that you have fast networking between the Tibco broker and your servers.
We just happened to find out that the hardware folks gave us legacy 100MB connections between our Tibco EMS broker and our CEP Input Server. We asked them to upgrade us to GigE immediately. Unless you have personally installed your hardware, never assume anything in your infrastructure.
2) It might be helpful to turn the "AutoAcknowledge" flag off in your Tibco session.
To do this, change
this.m_connection.CreateSession(false, SessionMode.AutoAcknowledge);
to
this.m_connection.CreateSession(false, SessionMode.NoAcknowledge);
Be careful about using No Acks. What is the impact to your CEP system if you happen to drop a message?
3) Implement threaded queues on the input from Tibco to your input server, and on the output from your input server to the CEP system.
If you need more performance, change the single threaded queue that reads messages from Tibco to a load-balanced multiple queues, each with its own thread. The load balancer can partition by Tibco topic, by ticker, or by OrderId.
Of course, you can always use the built-in adapters from your CEP vendor. But that may come at an extra monetary cost, so check with your CEP vendor to see if a certain input adapter comes free of charge. Also, the more of the CEP vendor's infrastructure you use, the more it ties you into the CEP vendor.
4) Try to do some pre-filtering of messages in your input server before passing them on to the CEP engine.
Recently, we were able to reduce the CPU usage of Coral8 by about 50% (according to our Coral8 guy) by filtering out FIX messages that we knew that Coral8 was not interested in.
5) Continually optimize your code. Just this morning, I did a code review of the input server, and I was able to suggest a change which reduced one hashtable lookup on each FIX message. I know that there are tons of other places in our code that can do with some optimization. If you need to parse FIX messages into C# objects, then do some investigation and benchmarking of FIX engines.
6) Use the Server Garbage collector
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
The London stock Exchange Crash and .NET/Windows
I have been on the receiving side of some good-natured ribbing inside of my company in the wake of the LSE crash and the suspected involvement of .NET and Windows. Our Complex Event Processing System, one of the most strategic projects in our company, runs totally on Windows 2003/64-bit and .NET 3.5. It is one of the highest-profile usages of .NET in one of the largest financial institutions in the world.
Microsoft has been very public is touting the fact that the LSE is a heavy user of .NET. But, from what I can tell, there hasn't been a peep out of Microsoft in defending their platforms in the wake of the LSE debacle. Speculation has pointed to problems with Cisco, problems with SQL Server 2000, problems with the code written by Accenture consultants, and problems with Sarah Palin.
I have been vocal on my blog in the past about the viability of Windows and .NET as a platform for real-time trading applications. I have never seen Microsoft come out with a definitive roadmap for using its platforms for realtime trading apps. No benchmarks against Realtime Linux. No attempts to through OPRA feeds at a Microsoft Server.
Sure, there is some good stuff going on at Microsoft with their grid efforts, but that effort will be the most beneficial to risk and analytics. Things like Velocity Object Caches and SSDS are a lot of fun. Excel is everywhere. But, the core operating system is Microsoft's bread and butter.
We need Microsoft to defend the LSE's use of Windows and .NET, and we need Microsoft to address viability of using its platform for low-latency, high-bandwidth trading apps.
At the Stac Council meeting in July, the companies who gave presentations about using their platforms for trading apps included IBM, Sun, and HP. Where was Microsoft? Why aren't they involved in the Stac Council?
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
Microsoft has been very public is touting the fact that the LSE is a heavy user of .NET. But, from what I can tell, there hasn't been a peep out of Microsoft in defending their platforms in the wake of the LSE debacle. Speculation has pointed to problems with Cisco, problems with SQL Server 2000, problems with the code written by Accenture consultants, and problems with Sarah Palin.
I have been vocal on my blog in the past about the viability of Windows and .NET as a platform for real-time trading applications. I have never seen Microsoft come out with a definitive roadmap for using its platforms for realtime trading apps. No benchmarks against Realtime Linux. No attempts to through OPRA feeds at a Microsoft Server.
Sure, there is some good stuff going on at Microsoft with their grid efforts, but that effort will be the most beneficial to risk and analytics. Things like Velocity Object Caches and SSDS are a lot of fun. Excel is everywhere. But, the core operating system is Microsoft's bread and butter.
We need Microsoft to defend the LSE's use of Windows and .NET, and we need Microsoft to address viability of using its platform for low-latency, high-bandwidth trading apps.
At the Stac Council meeting in July, the companies who gave presentations about using their platforms for trading apps included IBM, Sun, and HP. Where was Microsoft? Why aren't they involved in the Stac Council?
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
Wednesday, September 10, 2008
Automated Trading and CEP on News Feeds
Something for us all to think about if we ever use automated news in systematic trading or CEP.
Wall Street Journal article here.
“The damage was exacerbated by the growing use on Wall Street of automated programs that trigger stock trades without any human interaction,”
This caused a drop of $1B in the market share of United Airlines.
I wonder if the sentiment engines (like Reuters Sentiment, SemLab, Corpora) would have caught this?
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
Wall Street Journal article here.
“The damage was exacerbated by the growing use on Wall Street of automated programs that trigger stock trades without any human interaction,”
This caused a drop of $1B in the market share of United Airlines.
I wonder if the sentiment engines (like Reuters Sentiment, SemLab, Corpora) would have caught this?
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
Saturday, September 06, 2008
CEP and Shoplifting
Yesterday, I decided to drive the long way for a lunch with an ex-colleague of mine from Morgan Stanley. The local NPR public radio station (WNYC) has replaced the heavy metal station (WSOU) as my primary station in my car (sigh ... the joys of getting older ...), and I caught the last part of the Leonard Lopate show, a very popular news-talk show in New York.
Leonard was interviewing John Colapinto, who had written an article for The New Yorker on loss prevention in stores ... commonly known as shoplifting.
What was fascinating was that, as John was talking about how store detectives monitor customer patterns for shoplifting, it seemed like the perfect thing for Complex Event Processing (I hope that Tim will agree that this is truly Complex).
Now, I am sure that this is not news to the CEP vendors out there. Fraud detection and surveillance is one of the big applications of CEP. But, after each detection pattern Colapinto described, I said to myself "How would we do this in Coral8?".
Some of the patterns Colapinto mentioned included:
1) Customers examining clothing without looking at the price tag.
2) Customers moving randomly from table to table. There seems to be a certain pattern of movement that is common amongst shoplifters.
In the "Shoplifter Alert System" that we would build, there would be multiple levels of complex events (Tim can correct me on the proper terminology to use) :
Level 1 Event
"This person is a shoplifter"
Level 2 Events
"This person is moving too randomly through the store"
"This person does not seem interested in the price"
Level 3 Events
The amount of time a person is spending at each "table" in the store
The amount of time a person handles each garment
A sensor on the price tag to indicate whether the person has looked at the price
Level 4 Events
A person has entered the store
A person has left the store
A person is moving through the store
I would love to hear any anecdotes about how CEP is actually being used in detecting shoplifting, so if you know of anything, please comment.
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
Leonard was interviewing John Colapinto, who had written an article for The New Yorker on loss prevention in stores ... commonly known as shoplifting.
What was fascinating was that, as John was talking about how store detectives monitor customer patterns for shoplifting, it seemed like the perfect thing for Complex Event Processing (I hope that Tim will agree that this is truly Complex).
Now, I am sure that this is not news to the CEP vendors out there. Fraud detection and surveillance is one of the big applications of CEP. But, after each detection pattern Colapinto described, I said to myself "How would we do this in Coral8?".
Some of the patterns Colapinto mentioned included:
1) Customers examining clothing without looking at the price tag.
2) Customers moving randomly from table to table. There seems to be a certain pattern of movement that is common amongst shoplifters.
In the "Shoplifter Alert System" that we would build, there would be multiple levels of complex events (Tim can correct me on the proper terminology to use) :
Level 1 Event
"This person is a shoplifter"
Level 2 Events
"This person is moving too randomly through the store"
"This person does not seem interested in the price"
Level 3 Events
The amount of time a person is spending at each "table" in the store
The amount of time a person handles each garment
A sensor on the price tag to indicate whether the person has looked at the price
Level 4 Events
A person has entered the store
A person has left the store
A person is moving through the store
I would love to hear any anecdotes about how CEP is actually being used in detecting shoplifting, so if you know of anything, please comment.
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
Monday, September 01, 2008
Two strong .NET/C# Developers Wanted
We have openings for two .NET developers. One is for our options market-making operating in Jersey City. The other is for an FX trading system in New York.
1) No consultants. These are full-time positions.
2) Must have very strong experience in WinForms.
3) Real-time messaging experience needed. Sockets, Tibco (RV or EMS), MSMQ, MQ are all nice to haves.
4) Multi-threading experience is a must.
If you are interested, please send me your resume.
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
1) No consultants. These are full-time positions.
2) Must have very strong experience in WinForms.
3) Real-time messaging experience needed. Sockets, Tibco (RV or EMS), MSMQ, MQ are all nice to haves.
4) Multi-threading experience is a must.
If you are interested, please send me your resume.
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
Sunday, August 31, 2008
More on "Towards a Streaming SQL Standard"
The title of the presentation "Towards a Streaming SQL Standard" might be a little misleading, in my opinion. The title of the paper might have been "Oracle does this, Streambase does that, let's make a new language construct that enables both products to do this and that". And this is a great example of cooperation between vendors who researchers probably share a common academic heritage.
I don't pretend to understand why the new SPREAD construct is important. It will probably dawn on me when my team has an actual use case that the SPREAD construct solves. This is the case that I experienced a few months ago with Coral8 when we needed windows that flushed themselves when a certain column changed value (to be fair, Streambase already had this in their language).
A few points:
1) Despite what I think about Streambase's marketing and sales organization, you must admit that Zdonik and Cherniack are first-class researchers, and have contributed a lot to the field of CEP.
2) I get confused about Oracle's CEP offering. Is this paper talking about the CEP product that came with the BEA purchase or the original Oracle CEP product? Does any of this include work that they may have incorporated from the purchase of ESPER?
3) I would love to see Coral8's and Aleri's responses to this paper. Do their versions of Streaming SQL already do what the new SPREAD operator is purporting to do?
4) Will any of this standards work bubble up to the work that the STAC A1 council is doing? The STAC A1 council must be vigilant to ensure that we don't include benchmarks that might show off a certain, vendor-specific feature, unless this new feature solves an important business case. Likewise, if the SPREAD operator is important enough that all vendors rush to implement it, then this should be part of the STAC benchmark suite.
5) If the SPREAD operator is important, I expect that Coral8 will implement it right away.
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
I don't pretend to understand why the new SPREAD construct is important. It will probably dawn on me when my team has an actual use case that the SPREAD construct solves. This is the case that I experienced a few months ago with Coral8 when we needed windows that flushed themselves when a certain column changed value (to be fair, Streambase already had this in their language).
A few points:
1) Despite what I think about Streambase's marketing and sales organization, you must admit that Zdonik and Cherniack are first-class researchers, and have contributed a lot to the field of CEP.
2) I get confused about Oracle's CEP offering. Is this paper talking about the CEP product that came with the BEA purchase or the original Oracle CEP product? Does any of this include work that they may have incorporated from the purchase of ESPER?
3) I would love to see Coral8's and Aleri's responses to this paper. Do their versions of Streaming SQL already do what the new SPREAD operator is purporting to do?
4) Will any of this standards work bubble up to the work that the STAC A1 council is doing? The STAC A1 council must be vigilant to ensure that we don't include benchmarks that might show off a certain, vendor-specific feature, unless this new feature solves an important business case. Likewise, if the SPREAD operator is important enough that all vendors rush to implement it, then this should be part of the STAC benchmark suite.
5) If the SPREAD operator is important, I expect that Coral8 will implement it right away.
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
Saturday, August 30, 2008
4th Annual Event Processing Symposium
The schedule for the 4th Annual Event Processing Symposium has been posted.
Opher was kind enough to invite me to participate on the various panels in order to give the perspective from a financial services end-user. Unfortunately, I had to decline, as I can not really miss an entire week of work at this point (the first two days of the week will be spent at the Gartner Event Processing Summit, where unfortunately, I will not have the pleasure of ogling Tim Bass's beautiful wife).
I am disappointed to see that the customer rep from Financial services is Ian Koenig, formerly of Thomson Financial. There are a few factors that contribute to this disappointment:
1) Thomson is a vendor. It's not really a customer. It supplies products to trading firms like mine.
2) If I remember last year's Gartner Summit, Ian's forte was streaming news, not trading. (I hope that Brian Theodore might be able to clarify this a bit). Streaming news, while possibly useful in automated trader, is not something that is truly representative of the needs of a trading firm.
3) As it states on the programme, Ian is now an independent consultant. This means that, to me, he is not a true employee of the financial services industry. I would feel the same way if someone from Accenture or Sunguard Consulting marketed themselves as being an employee of the financial services industry .... but this is certainly my own feeling ... your mileage may vary.
(As an aside: When my old boss was dismissed from my firm last year, he continued to speak at conferences under the guise of still being a member of my company. I notice that this is a trend amongst people who have been recently dismissed from their financial services firms ... even on LinkedIn, people's profile still indicate that they are a member of a firm that they have been separated from.)
I get a lot of calls, asking me to speak at conferences, be on panels, be quoted for interviews, etc. I am happy to do so under various conditions. First, the appearance has to be approved by the senior communications people at my firm ... something that they have graciously granted so far. Second, I cannot reveal any of our "secret sauce".
I have a feeling that one of the reasons that Opher had to "fall back" on Ian is because most financial services firms are not as permissive as mine. I know most of the people involved in CEP on Wall Street, and I am sure that most of their firms would be dead set against them speaking at a public conference and blogging. We are a fairly secretive group, although in reality, all of us know what the others are currently working on. There are usually only two degrees of separation between any two people on Wall Street. That's why I am a bit puzzled as to why I seem to be the only customer from financial services who is willing to talk openly about CEP, and how excited we are to embrace the technology in our company.
On another note about the EPTS conference, the panel discussion titled "Is event processing a hype or the best invention since sliced bread ? Event processing market from business perspective" is chocked full of vendors. What is the blazes are vendors doing on this panel? Tim Bass ... where are you when we need you !
On a positive slant, there is a presentation by a Venture Capitalist on why he decided to stake some $$$ into a CEP company. That talk looks like it is worth the price of admission.
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
Opher was kind enough to invite me to participate on the various panels in order to give the perspective from a financial services end-user. Unfortunately, I had to decline, as I can not really miss an entire week of work at this point (the first two days of the week will be spent at the Gartner Event Processing Summit, where unfortunately, I will not have the pleasure of ogling Tim Bass's beautiful wife).
I am disappointed to see that the customer rep from Financial services is Ian Koenig, formerly of Thomson Financial. There are a few factors that contribute to this disappointment:
1) Thomson is a vendor. It's not really a customer. It supplies products to trading firms like mine.
2) If I remember last year's Gartner Summit, Ian's forte was streaming news, not trading. (I hope that Brian Theodore might be able to clarify this a bit). Streaming news, while possibly useful in automated trader, is not something that is truly representative of the needs of a trading firm.
3) As it states on the programme, Ian is now an independent consultant. This means that, to me, he is not a true employee of the financial services industry. I would feel the same way if someone from Accenture or Sunguard Consulting marketed themselves as being an employee of the financial services industry .... but this is certainly my own feeling ... your mileage may vary.
(As an aside: When my old boss was dismissed from my firm last year, he continued to speak at conferences under the guise of still being a member of my company. I notice that this is a trend amongst people who have been recently dismissed from their financial services firms ... even on LinkedIn, people's profile still indicate that they are a member of a firm that they have been separated from.)
I get a lot of calls, asking me to speak at conferences, be on panels, be quoted for interviews, etc. I am happy to do so under various conditions. First, the appearance has to be approved by the senior communications people at my firm ... something that they have graciously granted so far. Second, I cannot reveal any of our "secret sauce".
I have a feeling that one of the reasons that Opher had to "fall back" on Ian is because most financial services firms are not as permissive as mine. I know most of the people involved in CEP on Wall Street, and I am sure that most of their firms would be dead set against them speaking at a public conference and blogging. We are a fairly secretive group, although in reality, all of us know what the others are currently working on. There are usually only two degrees of separation between any two people on Wall Street. That's why I am a bit puzzled as to why I seem to be the only customer from financial services who is willing to talk openly about CEP, and how excited we are to embrace the technology in our company.
On another note about the EPTS conference, the panel discussion titled "Is event processing a hype or the best invention since sliced bread ? Event processing market from business perspective" is chocked full of vendors. What is the blazes are vendors doing on this panel? Tim Bass ... where are you when we need you !
On a positive slant, there is a presentation by a Venture Capitalist on why he decided to stake some $$$ into a CEP company. That talk looks like it is worth the price of admission.
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
Brief Thoughts on Standardized Streaming SQL
I haven't blogged for a few weeks because I am been on vacation for two weeks (San Francisco, Vancouver, Whistler, Banff) and in between that, I have been bogged down with all of the paperwork at my job, including doing budget planning for 2009. Let me tell you that, in this current economic environment, budget planning is a very interesting and challenging exercise. (The best strategy for dealing with budget time is to take your favorite managing director on the business side out to The Brandy Library, make sure he gets all liquored-up, and make him sign off on your budget after his 5th Highland Cooler.)
Also, I have been a bit turned off by all of the recent in-fighting that has been occurring on the CEP-related blogs. It seems that one person posts an opinion, then two or three of the well-known pundits start a counter-argument on their own blogs, which leads to a spiral of (sometimes good-natured) venom. Just like the recent Democratic and Republican conventions in the United States, I am hoping that the Gartner CEP Summit in two weeks will be a big love fest, and that the pundits will make peace with each other for a week or two.
I read with great amusement the recent announcement of Oracle and Streambase getting together and attempting to define a new standard for Streaming SQL. Some things that immediately crossed my mind were:
1) Streambase and Oracle are presenting their paper at the VLDB conference. The list of presentations is very impressive. Notice the number of papers that Microsoft is presenting, which leads me to hope that some very interesting stuff will be coming down the pike from MSFT. I wish that I could attend the tutorial on "Detecting Clusters in Moderate-to-High Dimensional Data: Subspace Clustering, Pattern-based Clustering, and Correlation Clustering".
2) Mark Palmer, who railed against Streaming SQL for such a long time in favor of Apama's more procedural language, now has to publically support the effort to standardize Streaming SQL. I wonder how much of Apama's language will appear in the standardized language.
3) There are other efforts at standardization underway. Opher Etzion is involved in one of them. I am not sure if Opher's efforts are aimed at solely defining a meta-language for events, or if what he is working on is in direct competition to the Streambase/oracle effort.
4) How willing will Aleri, Apama, and Coral8 be in adopting this effort? In particular, Aleri has a richer programming environment because of their procedural FlexStream language that gives developers a procedural "out" from the SQL-based language. Apama prides themselves on their Java-like language.
5) What happens if Microsoft ever weighs in with something of their own? You know that Microsoft will tie any effort in this area in with LINQ. Aleri is also moving to a more LINQ-like way of doing things. Of course, one can write a LINQ provider for Streaming SQL, but would anyone be motivated to do so?
(Update: Thanks to a reader who does not wish to be identified... the Streambase/Oracle paper is located here)
----------------------
On another note, good luck to Colin Clark, who has left Streambase as quickly as he joined them. Colin looks like he is striking out on his own as an indie consultant. Colin just had his first child, so this must be an especially interesting time for him. It reminds me a bit of my own story when , in 1988, I quit consulting for Goldman Sachs while my mortgage application was pending in order to strike out on my own. When you get the entrepreneurial fever, nothing can stop you ....
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
Also, I have been a bit turned off by all of the recent in-fighting that has been occurring on the CEP-related blogs. It seems that one person posts an opinion, then two or three of the well-known pundits start a counter-argument on their own blogs, which leads to a spiral of (sometimes good-natured) venom. Just like the recent Democratic and Republican conventions in the United States, I am hoping that the Gartner CEP Summit in two weeks will be a big love fest, and that the pundits will make peace with each other for a week or two.
I read with great amusement the recent announcement of Oracle and Streambase getting together and attempting to define a new standard for Streaming SQL. Some things that immediately crossed my mind were:
1) Streambase and Oracle are presenting their paper at the VLDB conference. The list of presentations is very impressive. Notice the number of papers that Microsoft is presenting, which leads me to hope that some very interesting stuff will be coming down the pike from MSFT. I wish that I could attend the tutorial on "Detecting Clusters in Moderate-to-High Dimensional Data: Subspace Clustering, Pattern-based Clustering, and Correlation Clustering".
2) Mark Palmer, who railed against Streaming SQL for such a long time in favor of Apama's more procedural language, now has to publically support the effort to standardize Streaming SQL. I wonder how much of Apama's language will appear in the standardized language.
3) There are other efforts at standardization underway. Opher Etzion is involved in one of them. I am not sure if Opher's efforts are aimed at solely defining a meta-language for events, or if what he is working on is in direct competition to the Streambase/oracle effort.
4) How willing will Aleri, Apama, and Coral8 be in adopting this effort? In particular, Aleri has a richer programming environment because of their procedural FlexStream language that gives developers a procedural "out" from the SQL-based language. Apama prides themselves on their Java-like language.
5) What happens if Microsoft ever weighs in with something of their own? You know that Microsoft will tie any effort in this area in with LINQ. Aleri is also moving to a more LINQ-like way of doing things. Of course, one can write a LINQ provider for Streaming SQL, but would anyone be motivated to do so?
(Update: Thanks to a reader who does not wish to be identified... the Streambase/Oracle paper is located here)
----------------------
On another note, good luck to Colin Clark, who has left Streambase as quickly as he joined them. Colin looks like he is striking out on his own as an indie consultant. Colin just had his first child, so this must be an especially interesting time for him. It reminds me a bit of my own story when , in 1988, I quit consulting for Goldman Sachs while my mortgage application was pending in order to strike out on my own. When you get the entrepreneurial fever, nothing can stop you ....
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
Tuesday, August 12, 2008
Career Choices - IB vs CEP Vendor
Which would you rather do? Work for a big Wall Street Investment bank, be guaranteed a good salary and (in good times), a decent bonus that might make you the envy of your neighborhood (unless you live in Short Hills, NJ or Greenwich, Ct), or work for a small, technically-savvy software company where you can be directly responsible for the bottom line?
This choice was recently faced by a colleague (let's call him T) at the big IB that I work for.
T is one of the top performance architects on Wall Street, and has personally been instrumental at improving the performance of several of our key systems. For various reasons, T felt that he wanted to explore some other opportunities, and within a few weeks, he had a few job and consulting offers. The fact that he could be so successful in his job search in this kind of market speaks volumes about his talent.
The position that T was going to accept was as a W-2 consultant with another large investment bank whose troubles have been in the news lately. T had never consulted before, so he had asked me a lot of questions about what it was like to be a consultant. Between the fact that T was new to consulting, only had a one-year contract, and was involved with an Investment Bank whose management was said to be a bit befuldled, I preferred to see T end up in another place. If this Investment Bank were to lay off additional consulting staff in the future, T would probably be one of the first ones to go. The Investment bank would have probably given T a small chunk of money to buy off the rest of the contract, and then T would have found himself back out on the street.
Coincidentally, a week before T was to resign from my firm, the CEO of one of the Complex Event Processing vendors contacted me to ask me if I knew anyone who would be available for a new position that his company had opened. Immediately, I thought of T. He was on his way out the door anyway, so I felt OK about referring him to the CEP vendor. Within two days, T had a third offer, and after another day, T inked the deal with the CEP vendor!
I spent quite a bit of time with T to help him through his array of choices. Almost every decision point favored joining the CEP vendor. The CEP vendor is certainly not a start-up, as they have been around for a few years and have about 60 people working for them, but it certainly has the feel of a start-up (sans the foozball table and loft in Silicon Alley).
The things that favored the IB were:
- The ability to interact with a large number of people and make a lot of new connections immediately. In consulting, your network is everything.
- The ability to work on a wide variety of trading systems, and to improve his domain knowledge.
- A great hourly rate
The negatives about consulting for an IB were:
- This particular IB has been particularly caught up in the credit crisis, and has recently dismissed a good number of employees, as well as a lot of contractors
- Being subjected to IB politics and processes
- The costs associated with being a consultant, which include paying 14% for Social Security (even though he was a W-2 of the consulting firm, they required him to pay the entire SS tax!)
The things that favored the CEP vendor were:
- Work in a start-up-type atmosphere where there should be very little politics and processes that get in the way of your productivity
- Work in a company of extremely smart rocket scientists that is very close to T's home
- In a pre-sales role, be able to directly affect the bottom line of the company and drive the future success of the company
- Have an opportunity to travel (well, some might look on that as a negative)
- Be able to meet a lot of different technologists from a lot of different kinds of companies. Since CEP appeals to hedge funds, there is an opportunity to gain some domain knowledge into hedge fund operations.
- If T is able to secure any kind of equity position in the company, then there is an opportunity to cash-in if the company gets purchased.
- Have direct interaction with the senior management of the company.
- Been involved in a technology that is visible in the Garter Hype Cycle
- The opportunity to be a jack-of-all-trades ... sales, technology, performance analysis, consulting
The things that were negatives about joining the CEP vendor were:
- Sales of infrastructure products are very difficult right now
- Deal with the very long delays in the procurement process that are endemic to every Investment Bank
- The travel schedule of a sales engineer might be very onerous, especially if you have a family
- Doing booth duty at trade shows
Did T choose wisely? I would be interested to hear your thoughts.
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
This choice was recently faced by a colleague (let's call him T) at the big IB that I work for.
T is one of the top performance architects on Wall Street, and has personally been instrumental at improving the performance of several of our key systems. For various reasons, T felt that he wanted to explore some other opportunities, and within a few weeks, he had a few job and consulting offers. The fact that he could be so successful in his job search in this kind of market speaks volumes about his talent.
The position that T was going to accept was as a W-2 consultant with another large investment bank whose troubles have been in the news lately. T had never consulted before, so he had asked me a lot of questions about what it was like to be a consultant. Between the fact that T was new to consulting, only had a one-year contract, and was involved with an Investment Bank whose management was said to be a bit befuldled, I preferred to see T end up in another place. If this Investment Bank were to lay off additional consulting staff in the future, T would probably be one of the first ones to go. The Investment bank would have probably given T a small chunk of money to buy off the rest of the contract, and then T would have found himself back out on the street.
Coincidentally, a week before T was to resign from my firm, the CEO of one of the Complex Event Processing vendors contacted me to ask me if I knew anyone who would be available for a new position that his company had opened. Immediately, I thought of T. He was on his way out the door anyway, so I felt OK about referring him to the CEP vendor. Within two days, T had a third offer, and after another day, T inked the deal with the CEP vendor!
I spent quite a bit of time with T to help him through his array of choices. Almost every decision point favored joining the CEP vendor. The CEP vendor is certainly not a start-up, as they have been around for a few years and have about 60 people working for them, but it certainly has the feel of a start-up (sans the foozball table and loft in Silicon Alley).
The things that favored the IB were:
- The ability to interact with a large number of people and make a lot of new connections immediately. In consulting, your network is everything.
- The ability to work on a wide variety of trading systems, and to improve his domain knowledge.
- A great hourly rate
The negatives about consulting for an IB were:
- This particular IB has been particularly caught up in the credit crisis, and has recently dismissed a good number of employees, as well as a lot of contractors
- Being subjected to IB politics and processes
- The costs associated with being a consultant, which include paying 14% for Social Security (even though he was a W-2 of the consulting firm, they required him to pay the entire SS tax!)
The things that favored the CEP vendor were:
- Work in a start-up-type atmosphere where there should be very little politics and processes that get in the way of your productivity
- Work in a company of extremely smart rocket scientists that is very close to T's home
- In a pre-sales role, be able to directly affect the bottom line of the company and drive the future success of the company
- Have an opportunity to travel (well, some might look on that as a negative)
- Be able to meet a lot of different technologists from a lot of different kinds of companies. Since CEP appeals to hedge funds, there is an opportunity to gain some domain knowledge into hedge fund operations.
- If T is able to secure any kind of equity position in the company, then there is an opportunity to cash-in if the company gets purchased.
- Have direct interaction with the senior management of the company.
- Been involved in a technology that is visible in the Garter Hype Cycle
- The opportunity to be a jack-of-all-trades ... sales, technology, performance analysis, consulting
The things that were negatives about joining the CEP vendor were:
- Sales of infrastructure products are very difficult right now
- Deal with the very long delays in the procurement process that are endemic to every Investment Bank
- The travel schedule of a sales engineer might be very onerous, especially if you have a family
- Doing booth duty at trade shows
Did T choose wisely? I would be interested to hear your thoughts.
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
Friday, August 01, 2008
Stac Council Meeting Recap
On July 30th, 2008, Stac Research had a general meeting of Stac Council members and associated vendors. Bob from Deutsche Bank was the generous host who managed to procure the Deutsche auditorium for the event.
There was a really good turnout given the fact that it was a hot summer's night in prime vacation season. Peter told me that about 175 people said that they would attend, and I would estimate the final figure at around 100 people.
The event was divided into 3 parts.
Stac had invited 10 vendors to give (exactly) 5 minute pitches on their products. The vendors that presented included IBM, HP, Sun, Solace, Tervela, Violin, Azul, Exegy, ServerEngines, and Bay Micro. All presented their solution for HPC on Wall Street. (Not a single mention of .NET that night .... where were you, Microsoft, and why aren't you on the Stac Council?)
There were two short presentations on the current state of the Stac M1 (Market Data) and E1 (Event Processing) working groups, given by Rob Wallos of Citi and yours truly.
Following the presentations was a cocktail hour, where there was actually a DeutscheBank labeled red wine! Everyone saw a lot of old Wall Street faces, ex-bosses, former colleagues, competitors, etc, and it was nice to reminisce with some of the folks who I haven't seen for years.
Several people from the large vendors approached me with questions about how they could join the Stac A1 working group. What it seems to me is that CEP still needs a lot of publicity, as people are not exactly sure about how to approach it. Only Ralph Frankel from Solace acknowledged the existence of CEP in his presentation, and offered a solution about how Solace can accelerate CEP-based apps. Solace is certainly a company that is on my radar, as it has been mentioned to me by several colleagues.
It's my opinion that, if a company really has a compelling HPC solution for Wall Street, membership in Stac is mandatory. Eventually, non-members will be conspicuous by their absence.
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
There was a really good turnout given the fact that it was a hot summer's night in prime vacation season. Peter told me that about 175 people said that they would attend, and I would estimate the final figure at around 100 people.
The event was divided into 3 parts.
Stac had invited 10 vendors to give (exactly) 5 minute pitches on their products. The vendors that presented included IBM, HP, Sun, Solace, Tervela, Violin, Azul, Exegy, ServerEngines, and Bay Micro. All presented their solution for HPC on Wall Street. (Not a single mention of .NET that night .... where were you, Microsoft, and why aren't you on the Stac Council?)
There were two short presentations on the current state of the Stac M1 (Market Data) and E1 (Event Processing) working groups, given by Rob Wallos of Citi and yours truly.
Following the presentations was a cocktail hour, where there was actually a DeutscheBank labeled red wine! Everyone saw a lot of old Wall Street faces, ex-bosses, former colleagues, competitors, etc, and it was nice to reminisce with some of the folks who I haven't seen for years.
Several people from the large vendors approached me with questions about how they could join the Stac A1 working group. What it seems to me is that CEP still needs a lot of publicity, as people are not exactly sure about how to approach it. Only Ralph Frankel from Solace acknowledged the existence of CEP in his presentation, and offered a solution about how Solace can accelerate CEP-based apps. Solace is certainly a company that is on my radar, as it has been mentioned to me by several colleagues.
It's my opinion that, if a company really has a compelling HPC solution for Wall Street, membership in Stac is mandatory. Eventually, non-members will be conspicuous by their absence.
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
Thursday, July 31, 2008
Berkeley Visualization Labs
The Berkeley Visualization Labs has a nice Java (grrr..) tool kit for visualization here.
In particular, their FlowMap visualization is a nice way to see order flow, as orders come in and get split up amongst different systems.
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
In particular, their FlowMap visualization is a nice way to see order flow, as orders come in and get split up amongst different systems.
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
Sunday, July 27, 2008
Lock-Free Queues and Market Data
Courtesy of Craig ....
Lock Free Queues
And, if you scroll down to the end of the blog posting, the author has links for other lock-free data structures.
According to Craig, this is ideal for implementing the market-data pattern.
(The author of the blog, Julian Bucknall, is the CTO of DevExpress. This might become my new favorite component vendor. Their Xtra PivotGrid is a pretty nice grid for viewing OLAP cubes.)
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
Lock Free Queues
And, if you scroll down to the end of the blog posting, the author has links for other lock-free data structures.
According to Craig, this is ideal for implementing the market-data pattern.
(The author of the blog, Julian Bucknall, is the CTO of DevExpress. This might become my new favorite component vendor. Their Xtra PivotGrid is a pretty nice grid for viewing OLAP cubes.)
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
Squiggles vs Resharper
Squiggles for C# will be available in Service Pack 1 of Visual Studio 2008.
http://www.infoq.com/news/2008/07/CSharp-Squiggles
The big question for me is how it will cooperate with Resharper. Will be have a race condition where Resharper and Squiggles do simultaneous error checking and fight to be the first one to notify us? How about the possible overlap of rules?
I hope that the VS team is aware of tools like Resharper, and is not working in the typical isolationist-mode that is prevalent in Microsoft.
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
http://www.infoq.com/news/2008/07/CSharp-Squiggles
The big question for me is how it will cooperate with Resharper. Will be have a race condition where Resharper and Squiggles do simultaneous error checking and fight to be the first one to notify us? How about the possible overlap of rules?
I hope that the VS team is aware of tools like Resharper, and is not working in the typical isolationist-mode that is prevalent in Microsoft.
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
Saturday, July 26, 2008
New York Times Article on Wall St. Pay Drop
http://www.nytimes.com/2008/07/26/nyregion/26pay.html
It's sobering to think that we are down to 6 major investment banks in New York City : JP Morgan, Morgan Stanley, Lehman, Citigroup, Goldman Sachs, and Merrill Lynch. You also have Bank of America, Wachovia, DeutscheBank, Barcap, and UBS among the secondary employers (ie: banks whose primary headquarters are not in New York). If Lehman goes down, then it leave 5 major IB's.
Back in the day, you had many many more firms that Wall Street IT professionals could hop between. The old adage was that you had to change companies every two years, with each move resulting in a salary increase and a bump in title. You had companies like Chase, Banker's Trust, Manny Hanny, Kidder Peabody, Drexel Burnham, Chemical Bank, Republic Bank, and many more. Then, you had all of the consulting firms like PWC, Ernst and Young, Anderson, Coopers and Lybrand, etc.
With all of consolidation that has gone on, plus companies who were forced out of business by the regulators, you are now more limited in terms of companies to go to. There are countless hedge funds in Connecticut, but that means you have to commute to Connecticut.
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
It's sobering to think that we are down to 6 major investment banks in New York City : JP Morgan, Morgan Stanley, Lehman, Citigroup, Goldman Sachs, and Merrill Lynch. You also have Bank of America, Wachovia, DeutscheBank, Barcap, and UBS among the secondary employers (ie: banks whose primary headquarters are not in New York). If Lehman goes down, then it leave 5 major IB's.
Back in the day, you had many many more firms that Wall Street IT professionals could hop between. The old adage was that you had to change companies every two years, with each move resulting in a salary increase and a bump in title. You had companies like Chase, Banker's Trust, Manny Hanny, Kidder Peabody, Drexel Burnham, Chemical Bank, Republic Bank, and many more. Then, you had all of the consulting firms like PWC, Ernst and Young, Anderson, Coopers and Lybrand, etc.
With all of consolidation that has gone on, plus companies who were forced out of business by the regulators, you are now more limited in terms of companies to go to. There are countless hedge funds in Connecticut, but that means you have to commute to Connecticut.
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
Tuesday, July 22, 2008
Welcome, Patric
My CEP team is happy to welcome Patric, a very experienced developer who specializes in UI design.
Our GUI is shaping up very nicely. We made the decision to go to WPF, and under Scott's leadership, the GUI is starting to come together nicely. We have been getting a lot of oohs and aahs on the trading floor, and the buzz that the GUI is generating will hopefully give us space on the traders' screens.
We have just about reached the point where people stopped wondering what our row on the trading floor has been up to. We are now starting to get requests from different groups, which is a position that we were hoping that we would be in. As many people have said about CEP, once you show people what you can get from a CEP infrastructure, your next set of problems is how to balance all of the requests that you will get.
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
Our GUI is shaping up very nicely. We made the decision to go to WPF, and under Scott's leadership, the GUI is starting to come together nicely. We have been getting a lot of oohs and aahs on the trading floor, and the buzz that the GUI is generating will hopefully give us space on the traders' screens.
We have just about reached the point where people stopped wondering what our row on the trading floor has been up to. We are now starting to get requests from different groups, which is a position that we were hoping that we would be in. As many people have said about CEP, once you show people what you can get from a CEP infrastructure, your next set of problems is how to balance all of the requests that you will get.
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
Saturday, July 19, 2008
The 2008 Waters Rankings
(If you are getting this post at the same time as you are getting my post about the Fluffies, read the Fluffies first.)
Coincidentaly, right after I posted the entry about the Jolt awards, I received my hardcopy of Waters Magazine in which they posted articles about their award winners.
What is the difference between the Jolt Awards and the Waters Rankings?
First, I trust the process that Waters has. In order to vote in the Waters Rankings, you need to be a subscriber to Waters. Most people in financial institutions can qualify for a free subscription to Waters, and you can fill out a form to apply. Waters will go through the trouble to qualify you for a subscription, and will make sure that you work for a financial firm. Unlike a magazine like Dr Dobbs, buying a copy of Waters magazine is difficult to do, and individual copies are quite expensive. Therefore, my conclusion is that Waters has a subscription base of highly-qualified financial professionals who have experience with at least some of the products that they vote for.
Second, in the Waters process, the readers themselves vote on the products. Waters received 600 votes this year. In fact, I am constantly receiving emails from Waters, encouraging me to vote on this and that. So, unlike a secretive panel of "highly renowned judges", the users of the products themselves are the ones who vote in the categories. (For the record, I do not participate in the voting.) Waters also goes through a process to weed out votes from the vendors themselves.
So, whereas the Jolt Awards have a lot of question marks associated with them, I trust the Waters Rankings more because I expect that the voters have actually evaluated or used the products that they vote for.
Just like Tim, I find the distinction between the Streaming Data Management category and the Complex Event Processing category to be a confusing one. Nevertheless, congratulations to Streambase's marketing machine. When I was evaluating Streambase, I found many good things about their product, and if their marketing and sales organization did not turn me off so much at the beginning, we might very well be using Streambase instead of Coral8.
It was also good to see Kx Systems in there as the number 3 solution. Kx certainly does not have the marketing oomph that Streambase has, relying on the mystique of Arthur Whitney and its Q and K languages instead. I would consider KDB to be more of a pure streaming data management solution (along with Vhayu) than a CEP solution.
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
Coincidentaly, right after I posted the entry about the Jolt awards, I received my hardcopy of Waters Magazine in which they posted articles about their award winners.
What is the difference between the Jolt Awards and the Waters Rankings?
First, I trust the process that Waters has. In order to vote in the Waters Rankings, you need to be a subscriber to Waters. Most people in financial institutions can qualify for a free subscription to Waters, and you can fill out a form to apply. Waters will go through the trouble to qualify you for a subscription, and will make sure that you work for a financial firm. Unlike a magazine like Dr Dobbs, buying a copy of Waters magazine is difficult to do, and individual copies are quite expensive. Therefore, my conclusion is that Waters has a subscription base of highly-qualified financial professionals who have experience with at least some of the products that they vote for.
Second, in the Waters process, the readers themselves vote on the products. Waters received 600 votes this year. In fact, I am constantly receiving emails from Waters, encouraging me to vote on this and that. So, unlike a secretive panel of "highly renowned judges", the users of the products themselves are the ones who vote in the categories. (For the record, I do not participate in the voting.) Waters also goes through a process to weed out votes from the vendors themselves.
So, whereas the Jolt Awards have a lot of question marks associated with them, I trust the Waters Rankings more because I expect that the voters have actually evaluated or used the products that they vote for.
Just like Tim, I find the distinction between the Streaming Data Management category and the Complex Event Processing category to be a confusing one. Nevertheless, congratulations to Streambase's marketing machine. When I was evaluating Streambase, I found many good things about their product, and if their marketing and sales organization did not turn me off so much at the beginning, we might very well be using Streambase instead of Coral8.
It was also good to see Kx Systems in there as the number 3 solution. Kx certainly does not have the marketing oomph that Streambase has, relying on the mystique of Arthur Whitney and its Q and K languages instead. I would consider KDB to be more of a pure streaming data management solution (along with Vhayu) than a CEP solution.
©2008 Marc Adler - All Rights Reserved.
All opinions here are personal, and have no relation to my employer.
Subscribe to:
Posts (Atom)
