![]()
©Celoxica 2008. T: +44 (0)1235 863656
Celoxica's Accelerator for Market Data Capture is the first product on the
market to offer ultra-low-latency access to market feed data. With average latency figures below 2 microseconds, Celoxica is redefining the term “low latency”.
Celoxica leverages hardware acceleration to provide an innovative new architecture for access to market data feeds. This architecture gives high reliability during spikes in message volume. Integrated into standard, tier 1 OEM servers, the Celoxica Accelerator for Market Data Capture offers the ultimate combination of performance and ease-of-use.
A co-processor directly connected to the network port means that the Accelerator can provide which is independent of the data volume. The HyperTransport link to the host CPU provides the lowest latency bus transfers available in standard servers.
The innovative architecture of the Accelerator means that it can deal with fluctuating, high message volumes.
The low-power co-processor architecture means that even when processing two fully-saturated 1 Gbps network links, the total power consumption of the server is barely affected.
Simple integration into existing applications on standard servers, with the ability to customise the message interface.
Hardware co-processors have been discussed for a while in high-performance computing circles as a method for
dramatically increasing system performance and offloading computation from CPUs. Celoxica's Accelerator is the first market data handler to directly connect a network interface to a co-processor, eliminating one of the major contributors to latency in a hardware/software co-processing system: the peripheral bus transactions between the co-processor and the network device.
Once in the co-processor, the network data is processed in a highly-parallel processing pipeline at line-speed. The TCP/UDP/IP protocol stack, FAST decoder, recovery system and ASCII-to-binary conversion engine are all implemented in dedicated hardware, with the lowest-possible latency.
The co-processor board fits into a standard HyperTransport slot, which is the lowest latency server communications bus available today. Once converted to a binary representation, the network data is transferred via the HyperTransport
bus directly to the CPU's memory. A C/C++ API is then available for you to integrate this system with your own application, getting feed updates before the competition.
For each market, the Accelerator is configured with efficient co-processing logic tailored to that market feed.
Celoxica recognises the importance of fitting in to your organisation's needs, so we give a high-degree of control over the behaviour of the Accelerator.
The Accelerator Card fits into the slot of an industry standard 1U rack-mount server. With an Accelerator, each server can process the full data rate of one market feed. Multiple Accelerators and servers can be deployed to process different markets.