See, Understand & Optimize Business Delivery Over Complex Infrastructures
Many factors affect trading service performance, leading to frustrated traders and clients, reduced order flow and ultimately lost revenues.
Corvil provides end-to-end visibility of every transaction by tracking it across Gateways, SORs, network components and market venues. Corvil’s multi-hop analysis quickly pinpoints performance degradations leading to swift resolutions and proactive optimizations that improve customer service and protect revenues.
– Director, Investment Bank
We have seen an increase in productivity by using Corvil. At least when it comes to the people who were put to task when there was an issue - who had to set up the sniffers and try to record a session and analyze packets - it was a pretty manual task. That has improved drastically. Now, with the same number of people, we are able to monitor more sessions in real-time.
It’s difficult to know if you’re really getting the bandwidth you’re paying for. Corvil’s Microburst Analysis provides the answer.
Poorly balanced load can have an impact on performance, Corvil shows how each gateway is loaded along with individual performance.
Corvil’s Multi-hop Analysis automatically tracks the path an order takes, revealing the latency across each hop.
Are the clients with the highest message rates also the ones who generate the highest revenue? Corvil provides the numbers.
Corvil’s patented technologies measure the per-packet latency between sites. It also relates latency performance to microburst activity, identifying if issues are due to the service provider or traffic congestion.
If you’d like to understand how quickly your trading gateway is processing clients orders, Corvil provides the answers, by tracking the latency of every message, in both directions.
By analyzing packets either side of an application such as an order gateway or pricing engine, Corvil can calculate the latency for each message processed by the application.
Corvil’s sophisticated correlation engine and extensive decoder suite handles the tracking of messages across different protocols.
By simultaneously monitoring packets either side of a network hop, Corvil’s correlation engine reports the one-way latency for every packet.
Within a Data Center, correlations are typically performed on a single Corvil appliance to measure the performance of firewalls and load balancers.
However correlations may also be performed between Corvil appliances in different physical locations to measure the latency of service providers and wide area network links. In the absence of a GPS or PTP source Corvil also takes care of clock synchronization.
By analyzing the microburst activity of traffic received across a link, Corvil reveals the effective bandwidth delivered by the service provider and confirms whether or not you’re getting what you paid for.
By monitoring both sides of a provider link with Corvil, it allows packet loss to be detected and the one-way latency of every packet to be measured. Corvil also takes care of clock synchronization supporting PTP & GPS/PPS. Plus, Corvil’s proprietary algorithm provides clock sync in the absence of an external source.
When a client order experiences a high response time, it’s important to understand precisely where the latency was introduced, so that remedial action can be taken.
With a single click, Corvil’s Multi-hop analysis, tracks a transaction through each hop of the infrastructure, including the associated response message in the reverse direction.
The results are displayed in a graphical sequence diagram, detailing the precise latency for each hop.
When delivering bursty data across Wide Area Network links, Microbursts can cause queuing and delays at the sending end.
Corvil models this queuing effect in real-time, revealing not only the estimated queuing latency but also providing bandwidth recommendations.
Corvil enables you to right size your network bandwidth to ensure the queuing latency (and data lag) doesn’t exceed your desired SLA.
Corvil’s microburst technology continuously analyzes the arrival of every packet from a network link to reveal any bandwidth bottlenecks that the traffic has encountered. It does this by detecting the smoothing effect that a network buffer has on bursty traffic.
In contrast, traditional network monitoring tools that report to a granularity of one second or above are blind to these microbursts, providing a misleading and over-optimistic view of network performance.
Recorded with UTC synchronized, nanosecond time-stamps, Corvil provides a highly granular and actionable audit trail of all activity, accurately time sequenced to understand cause and effect.
See infrastructure latency in true real-time with flexible and intuitive dashboards. Explore and interact with the data. Quickly drill into events to investigate root cause, or select longer time periods to spot trends, performance patterns & anomalies.
Corvil’s correlation engine reports the per-message latency across each infrastructure hop. It takes care of multiple protocols and messaging changes that occur across devices such as Gateways and SORs.
Using Microburst rather than average measurements, Corvil provides a detailed understanding of transient and dynamic loads, enabling infrastructure components to be correctly scaled to meet performance requirements.
Corvil’s multi-hop analysis tracks transactions across network and application hops, to execution venues and back again. It provides an intuitive and graphical representation of the latency over the entire journey.
Analysing transactions in real-time reveals performance anomalies as they happen. When latency spikes occur, Corvil immediately notifies operations teams and provides the crucial data to understand and resolve the event.