The case is that if you are a buy-side company then you have to use an algorithmic trading because everybody does; if you are a sell-side, the situation is the same, to make a huge deal you need to use an algorithm for the better price in better time-frame. Hope you have already heard about innovations and nanotechnologies.
Yes, I am also looking for this trend; and it appears in messaging engines. TIBCO presented TIBCO FLT which sets 2 new performance world records – an average application latency of 384 nanoseconds for intra-host communication and 3.1 microseconds for inter-host communication [1]. That’s about 40% faster than the nearest competitor.
In Luxoft we have TIBCO EMS and TIBCO Rendezvous expertise which is focused on investments banks, so the appearing of TIBCO FCL is a good opportunity for us and our major clients. Mostly, it is interesting to compare it with Informatica (29West) Ultra Messaging solution in work. Maybe someone wants to see the results?
[1] TIBCO FTL Benchmarks Performance metrics based on testing conducted by TIBCO in March 2011. Test cases measured average one way end-to-end application latency based on round trip times for messages. This can be replicated using sample applications tibping.c and tibpong.c available as a part of the TIBCO FTL® version 1.0.0 product download. Shared memory performance claim based on testing done using TIBCO FTL Version 1.0.0 running on a machine with Intel Xeon 5680 CPU, 48 GB RAM and Linux 2.6.35. RDMA over InfiniBand performance claim based on testing done using TIBCO FTL Version 1.0.0 running on a machine with Intel Xeon 5660 CPU, 40 GB RAM and Linux 2.6.18.
