Friday, November 16, 2007

Preparing for unexcepected events - pickpocket


Today I have experienced an unexpected event - I have been most likely pick pocketed while unloading to the car bags from shopping at the supermarket. When I noticed it - it was too late, the wallet was gone. It will take me some time and effort to reconstruct its content - had to cancel credit cards, get substitute for driving license, get new identity card, and some other less important cards as well. Event processing can help in this case -- first locating chip attached to driving license, credit cards etc - can help locate them (for the privacy fans - this will also let some big brother trace you), some automatic procedure to replace everything that has been there, without bothering me - would also been helpful. This bring the question of cost-effectiveness - the investment in handling relatively rare events which have a noticeable overhead vs. resolution when they do happen - this is a utility function which weights the cost, other damages, and the damage from the event that happens - an interesting investigation area - which also has practical impact on applying event processing in business situations -- which events should be traced and handled, and which not.

And since the tone of this blog is personal this time, I would like also to write a few lines about a good friend and colleague - Shlomit Zak - who unfortunately passed away recently in relatively young age. Many of the foundations of the work I have done over the years came from my joint work with Shlomit in the Israel Air-Force, many years ago, Shlomit has been one of the brightest persons I know, who, unfortunately did not come close to fulfill her potential; besides her professional abilities she has has an artistic soul, was gifted piano player, and keen sense of humor, and was famous by writing achrosticons. My last Email from her was two months ago when she sent me an Email "happy new year" card with achrosticon - and I have replied : "happy new year to you too - the achrosticon is nice, as usual; rhyming is mostly reasonable", and got the answer "and you - as usual, cannot free yourself from the need to provide grades" (well in Hebrew it sound better...).

Thursday, November 15, 2007

The MARK on the BENCH - and the mythical event per second

















Recent news item from BEA is talking about a benchmark and cites some EPS (Event Per Second) figures. Unlike some vendors that just cite numbers, there is also a white paper describing the benchmark. I don't wish to refer to the BEA benchmark specifically, but to share some insights about benchmarks in general. Benchmarks have a positive side, in which they are enable either to compare different products based on the same criteria, or to evaluate some properties of a product, even when not comparing it to others. Currently there is no "standard" benchmark in the event processing area, thus, vendors are inventing their own benchmark, carefully designed to expose much of the strengths, and none of the weaknesses of their products, and create benchmarks that may be non reproducible in other environments, or with some change in the application. Thus, to make any significant comparison between different products, standard benchmarks need to be constructed. Standard benchmarks, by themselves, may be double-edge sword, since we have benchmark-driven industry, vendors will invest a lot of resources into optimizing for the standard benchmark, however - this may not help a specific application, since its requirements may be far enough from the benchmark. Event Processing is heterogeneous area, which means that a single benchmark will not be sufficient - we need a collection of benchmarks, and each customer will have to chose the one or more benchmarks that are closer to its requirements. The standard benchmark should come from a vendor-neutral organization. I know of some academic work in this area, but more needed to be done.
And a word of caution - all the benchmarks refer to performance characteristics such as latency and throughput. But as noted in a previous post on the mythical event per second, I doubt if these are the main decision criteria in most applications - thus benchmarks should refer to other dimensions (functions, consumability, other non functional requirements), while, there are certainly cases that the high performance characteristics are critical, in general, I think this is over-hyped a bit. more - later.




















Tuesday, November 13, 2007

On more or less important issues - the case of event processing languages




There are things important in life - and things that are less important.
In the picture you can see my four daughters ride two elephants in our family vacation in Thailand earlier this year, for me they are obviously very important, while other things may be less important in life. I am returning to the issue of event processing languages, since I realize that too much are being spent over the community on comparison of what a certain language syntax can or cannot do - I admit that I have also contributed my share to this discussion.
Now I am raising the question --- is it an important topic to deal with? yes - I realize that some vendors would like to make a language as a differentiator (I am not sure if anybody succeeded in this task), it is also true that some people in the development community have programming style preference, but since I see the trend of getting the patterns/rules/queries being constructed by semi-technical people, who will not be able to use many of the languages anyway, and will need higher level abstractions anyway.
Therefore, I think that the more important thing to understand are - requirements and semantic features of event processing - so I intend, at least for a while, to stay away from the programming style and concentrate more on these two issues.
more - later.