On 24 Sep 2008 in LL-Data Fabrics, LL-Feed Handlers, LL-Multi-Core, Low-Latency.com, MDI-Data Feed Handlers, Market Data Insight
NYSE Euronext Advanced Trading Solutions, the low latency trading technology division of NYSE Euronext, has rolled out a software based solution to handle 10 major US equities feeds on a single server with potential headroom of three-times the current market peak capacity. The key ingredients of this implementation are its, Wombat Data Fabric, its high performance middleware and its feed handler technology.
READ THE REST
On 21 May 2008 in LL-Hardware Acceleration, LL-Multi-Core, Low-Latency.com
Azul Systems launched its Vega 3 Series appliances, designed to run, process, and manage Java applications in business critical environments. Compared to the company’s first generation devices, Vega 3 products represent a fivefold improvement in performance, packaged in compact 5U or 14U appliance.
READ THE REST
On 27 Feb 2008 in LL-Benchmark Initiatives, LL-Multi-Core, Low-Latency.com
Intel recently released quad-core Xeon processors (code named “Harpertown”) based on a 45nm manufacturing process, which the company claims significantly reduce power consumption compared to its previous generation of quad-core processors (”Clovertown”). Securities Technology Analysis Center (Stac) performed a study comparing the power consumption of a server using Harpertown processors to the same server using Clovertown processors, running a market data load of 2 million updates per second across a stacked RMDS configuration that utilised as much of the eight cores in this system as possible. The Harpertown processor decreased the server’s overall power consumption by 16 per cent (333 watts versus 395 watts) and increased the server’s market data efficiency (updates per second per watt) by 19 per cent (6010 ups/W versus 5060 ups/W).
Stac News Release 
Technorati Tags: Securities Technology Analysis Center, Stac, Intel, Intel Xeon Harpertown processor, Clovertown, RMDS
On 29 Jan 2008 in LL-Benchmark Initiatives, LL-Multi-Core, Low-Latency.com
Sun has joined the Securities Technology Analysis Center (Stac) Benchmark Council to help define standard performance metrics for trading technologies and has become an Innovation Sponsor for the Manhattan-based Stac Lab. On the Stac Benchmark Council, Sun will contribute to specifications under consideration by its participating trading firms. As an Innovation Sponsor, Sun is providing the lab with several of its latest technologies, including the 2U Sun Fire X4450 server with four quad-core Intel Xeon processors, the Sun Fire X4600 server with 8 AMD Opteron processors, and the 64-thread Sun T5120 server with UltraSPARC T2 processors and integrated 10 Gigabit Ethernet controllers.
Stac News Release 
Technorati Tags: Sun, Securities Technology Analysis Center, Stac, Stac Benchmark Council, Stac Lab, 2U Sun Fire X4450, Intel, quad-core Intel Xeon processors, Sun Fire X4600, AMD, AMD Opteron, Sun T5120, UltraSPARC T2, 10 Gigabit Ethernet
On 14 Jan 2008 in LL-Benchmark Initiatives, LL-Feed Handlers, LL-Multi-Core, Low-Latency.com
Quanthouse’s Quantfeedhandler, its feed handler technology to standardise exchange raw market data feeds, is able to decode more than two million messages per second, benchmark tests carried out at fasterLAB, the Intel low latency lab in London, on several versions of the multi-core Intel Xeon platform, have shown.
Technorati Tags: Quanthouse, QuantFeedhandler, Intel, Multicore Intel Xeon Processor, fasterLAB, exchange market data feeds, benchmarks
On 12 Dec 2007 in LL-Connectivity/Networking, LL-Data Fabrics, LL-Multi-Core, Low-Latency.com
Low latency technology provider Wombat is running a fully subscribed pilot of a new middleware solution which it claims enables ultra high performance messaging with minimal data centre resource consumption.
The vendor reckons it has proven the power of Wombat Data Fabric and its ability to provide market-leading density on open, standard x86 hardware by running a full data from OPRA, the US options market, on a single Intel-based server available in a single rack unit (1U) form factor. In the test using a full day of OPRA data, the OPRA feed handler, running on a two socket eight core Intel 5400 Xeon server, published all updates on all symbols to MAMA clients across a single Voltaire network switch. Average CPU utilisation was just 10 per cent, escalating to less than 35 per cent at peaks, with a mean latency of less than 100 microseconds. No speciality or proprietary hardware devices, such as FPGA cards, are required by the solution, reducing cost and complexity when compared with alternative “appliance” approaches, Wombat reckons.
Wombat News Release 
Technorati Tags: Wombat, Wombat Data Fabric, OPRA, Intel, Intel 5400 Xeon server, CPU utilisation, Voltaire, Voltaire network switch
On 04 Dec 2007 in LL-Benchmark Initiatives, LL-CEP/Event Stream Processing, LL-Multi-Core, Low-Latency.com
BEA Systems has published a benchmark of its CEP offering - Weblogic Event Server - demonstrating microsecond-level latencies when processing one million events per second. The benchmark was run on Intel’s Caneland platform, compring four quad core processors.
READ THE REST
On 29 Nov 2007 in LL-Benchmark Initiatives, LL-Multi-Core, Low-Latency.com
Aspeed Software is the first of Intel’s financial markets software partners to publish performance figures leveraging Intel’s recently introduced Xeon 5400 chip. Benchmarking a single-threaded options pricing application, Aspeed reports that a 5400-based system delivers a 2.2-fold run time reduction while consuming just 13 percent more power, when compared to a Xeon 5100-based system.
READ THE REST
On 25 Nov 2007 in LL-Connectivity/Networking, LL-Data Fabrics, LL-Events, LL-Multi-Core, Low-Latency.com
InfiniBand specialist Mellanox is hosting a series of seminars entitled “Accelerating the New Global Trading Markets” in Tokyo, Hong Kong, and Singapore in early December. They will focus on today’s financial applications and the latest technology for accelerating performance of trade execution and bandwidth intensive financial applications.
READ THE REST
On 22 Nov 2007 in LL-Benchmark Initiatives, LL-Multi-Core, Low-Latency.com
Intel Corporation has joined The Securities Technology Analysis Center (Stac) Benchmark Council and has entered into a multi-year lab agreement with Stac as an innovation sponsor. The agreement builds upon a benchmarking relationship that Intel established with Stac over a year ago. Intel has also made a multi-year commitment to provide the Manhattan-based Stac Lab with the latest Intel technologies. This includes its just-announced 45-nanometer Intel Xeon processor 5400 series (Harpertown), which preliminary Stac testing has found to decrease power consumption by 20 per cent over a previous Xeon processor 5300 series (Clovertown) when supporting the same market data workload.
Stac News Release 
Technorati Tags: Intel, Securities Technology Analysis Centre, Stac, Stac Benchmark Council, Stac Lab, Intel Xeon Processor 5400, Intel Xeon processor 5300, Harpertown, Clovertown
On 12 Nov 2007 in LL-Multi-Core, Low-Latency.com
Intel yesterday announced its revolutionary Xeon 5400 chip - the first based on its Penryn technology - and expects servers based on the device to deliver significant performance and power consumption benefits across a wide range of low latency financial applications, including risk analytics, market datafeed handling and transactional execution systems, according to Intel executives.
READ THE REST
On 01 Oct 2007 in LL-Benchmark Initiatives, LL-Multi-Core, Low-Latency.com
Intel and Rogue Wave Software have released their latest benchmark for a Hydra-based transactional application running on Intel’s Xeon 7300 Quad Core processors. A headline 68,940 transactions per minute was acheived. Hydra allows single threaded applications to leverage the benefits of multi threading, in a multi core environment.
Download Benchmark Overview 
Technorati Tags: intel, rogue wave, hydra, multi core
On 19 Sep 2007 in LL-Connectivity/Networking, LL-Messaging Middleware, LL-Multi-Core, Low-Latency.com
Cisco Systems has created a low latency datafeed offering, collaborating with Reuters, Sun Microsystems and Intel. The offering consists of the Reuters Market Data System installed on Sun servers running Solaris 10, powered by Dual-Core Intel Xeon processor 5100 series with Cisco Infiniband Server Fabric switches. Tests conducted by Sun acheived a maximum throughput rate of two million updates per second on a single Reuters Market Data System process.
Cisco News Release 
Technorati Tags: cisco, cisco systems, reuters, intel, sun, sun microsystems, solaris, infiniband
On 16 Sep 2007 in LL-Benchmark Initiatives, LL-Connectivity/Networking, LL-Messaging Middleware, LL-Multi-Core, Low-Latency.com
Stac has released a benchmark of Reuters Market Data System (RMDS 6) running on a single Intel “Caneland” server with four Quad Core Intel Xeon X7350 processors (for a total of 16 cores), running Red Hat Enterprise Linux 4.5 and communicating over Gigabit Ethernet.
READ THE REST
On 16 Sep 2007 in LL-Benchmark Initiatives, LL-Feed Handlers, LL-Messaging Middleware, LL-Multi-Core, Low-Latency.com
Stac has released a benchmark focused on mininizing transport latency, comprising Wombat Financial Software’s market datafeed simulator, 29West’s Latency Busters Messaging, a Sun Microsystems Sun Fire X4200 M2 with 2 Dual Core AMD Opteron 2.8 GHz processors and Solaris 10 update 3. The solution was able to deliver an overall mean latency of 84 microseconds at 50,000 updates per second over standard Gigabit Ethernet.
Stac News Release 
Technorati Tags: stac, securities analysis research center, wombat financial software, wombat, 29west, sun, sun microsystems, amd
On 11 Sep 2007 in LL-Benchmark Initiatives, LL-Connectivity/Networking, LL-Events, LL-Messaging Middleware, LL-Multi-Core, Low-Latency.com
Intel, Sun Microsystems, Reuters and Cisco Systems are leveraging Monday’s High Performance on Wall Street event to announce the results of tests on a high performance solution to accelerate financial services applications. Conducted by Sun, the tests included a Reuters application running on Intel, Sun, Reuters and Cisco technology, along with new Cisco high-performance software.
Visit www.highperformanceonwallstreet.com for more information and to register.
Technorati Tags: intel, sun, sun microsystems, reuters, cisco, cisco systems
On 16 Jul 2007 in LL-Algorithmic Trading, LL-Benchmark Initiatives, LL-CEP/Event Stream Processing, LL-Multi-Core, Low-Latency.com
London-based Datastream Analysis Ltd (DSAL) - a Streambase Systems partner and application developer - has benchmarked calculations used in algorithmic trading at 200,000 calculations per second and 1.5 million updates per second, leveraging Intel’s Low Latency Laboratory in the UK.
READ THE REST
On 10 Jul 2007 in LL-Benchmark Initiatives, LL-Hardware Acceleration, LL-Monitoring Tools, LL-Multi-Core, Low-Latency.com
Securities Technology Analysis Center (Stac) has released results of a benchmark conducted on Exegy’s Ticker Plant (XTP), showing 80 microsecond mean latency at one million updates per second. The benchmark configuration included the Field Processor Gate Array (FPGA)-based XTP version 1.2, Mellanox Infiniband and clients running Red Hat Enterprise Linux 4.4 on AMD multicore processors.
READ THE REST
On 25 Jun 2007 in LL-Data Fabrics, LL-Multi-Core, Low-Latency.com
Launched at last week’s SIFMA show, Gemstone Systems debuted Gemcache, comprising its Gemfire data fabric, running on IBM System x bladeservers, themselves based on Intel Quad-Core Xeon processors. The solution delivers high speed data distribution layered on I/O Acceleration technology and a high availability model and system management framework.
Technorati Tags: gemstone, ibm, intel, gemcache, gemfire, sifma
On 25 Jun 2007 in LL-Benchmark Initiatives, LL-Connectivity/Networking, LL-Messaging Middleware, LL-Multi-Core, Low-Latency.com
Reuters, Intel, HP and Voltaire are collaborating to pitch what they say is the lowest latency offering based around Reuters Market Data System (RMDS). The Infiniband-based tuned configuration reduces the overall system latency by as much as 48 percent (compared to an ethernet-based configuration), according to independent testing by the Securities Technology Analysis Center (Stac).
READ THE REST