TipOff enables capital market participants to effectively measure and manage performance in extremely fast moving and high volume environments, pinpointing opportunities for improvement. Capturing, decoding and normalising 100% of data received, TipOff provides real-time and retrospective analysis with access to detailed network, middleware and market data insight.
TipOff’s precise instrumentation enables performance issues to be forensically analysed, right down to the individual network packet and nanosecond level, so an issue’s root cause can be quickly identified and eliminated, facilitating improved performance, faster trade execution and accelerated matching speeds. Captured data can also be fed directly into client specific analysis tools providing vital risk and regulatory insight and, within pre-production environments, TipOff enables the market data received on any given day to be retransmitted, facilitating accurate and robust testing for new algorithms and system enhancements.
Available on an open platform, with standard APIs, TipOff can be rapidly deployed as both an Appliance and software only solution on physical and virtual machines.
Passive Latency Monitoring
TipOff supports passive latency monitoring of market data and transaction flows. By decoding packet streams in real time through all layers from network up to application message content, TipOff can monitor both packet and message latency, accounting for retransmission latency in the reliability layer. A single multi-hop flow may comprise several different middleware stacks or protocols. TipOff supports flows that undergo transformations through order aggregation, trading engines and execution venues.
Latency Root Cause Analysis
TipOff supports latency root cause analysis by leveraging its thorough analysis of the middleware stacks that comprise each hop in a data flow. The most significant source of latency is retransmission latency resulting from lost packets. TipOff monitors the internal operation of reliability layers within protocols such as TCP, RV, LLM, LBM and RRCP, providing protocol specific statistics and early warnings of operational conditions that result in increased latency.
TipOff provides powerful network analysis capabilities including support for 1/10/40 GE interfaces, packet capture to disk with 20 Gbit sustained performance, nanoburst analysis at OSI layers 1, 2 and 3, TCP retransmission and window exhaustion stats, TCP ack latency and packet layer latency analysis, all with precise timing to 10ns resolution.
Precise Timed Replay™
TipOff’s Precise Timed Replay capability enables TipOff to replay previously captured data, reproducing the original packet timing with high precision, including line rate bursts. Replays can be scheduled at the original recorded rate, a multiple or fraction of the original rate, or at a fixed packet rate. This capability is useful for infrastructure performance benchmarking, trading algorithm back testing and tick database population.
TipOff G6 Turbo
The latest TipOff G6 hardware, the sixth generation, has unrivalled performance, leveraging the very latest Intel CPUs coupled with both FPGA and ASIC based hardware acceleration. TipOff G6 is available with the Turbo option - a high performance database that enables TipOff to correlate and store up to one million transactions per second on a single appliance. TipOff G6 Turbo achieves a twenty fold (x20) performance gain over TipOff G5 for typical transaction latency monitoring use cases.