In April 2015, Velocimetrics 6.0 was released. This enhanced version features improvements included in response to client requests for expanded functionality to meet new business requirements, to further improve the user experience and address emerging market needs.
Please find below a summary of the major enhancements included within this release:
Enabling complex and high volume environments to be accurately analysed
Velocimetrics has always excelled in enabling complex environments, trading strategies or business flows to be monitored with uncompromised visibility, facilitating real-time alerting and instant access to detailed performance analytics.
However, historically there has been a natural limit to the volume of events that could be effectively monitored and analysed at the same time. These limitations are due to the restrictions of CPU, memory and the levels of persistence that can be achieved using a single machine.
With Velocimetrics 5.0 a clustered architecture was introduced, enabling Velocimetrics to scale over multiple servers and process significantly larger volumes of events in real-time than had previously been possible. The following diagram shows how the clustering approach works
This approach proved very effective, however it did involve setting up rules at the agent level as all events that needed to be correlated together required being sent to the same partition. Additionally, in 5.0, whilst the clustering approach offered significant alerting and analysis capabilities, opportunities were identified to improve individual item alerting and the analysis of distributions across all captured data.
With Velocimetrics 6.0 the methodology has been substantially refined. The partitions are now in constant communication with each other, removing the need to set up user-defined routing rules. Clustering is more or less transparent in Velocimetrics 6.0, enabling the highly scalable solution to be deployed with minimal effort. Additionally, all analysis limitations have been eradicated. Now all variations of analytical analysis and alerting possible with standard Velocimetrics is available in the clustered edition.
Significantly improved persistence
With Velocimetrics 6.0 persistence has been improved by an order of magnitude. This improvement is independent of whether Velocimetrics is running as a standalone server, or operating as a partition within a clustered deployment.
As every partition writes to an independent database instance, persistence can be scaled linearly by adding partitions to the cluster.
Enriched on-demand and scheduled reporting capabilities
Velocimetrics 6.0 offers significant management reporting capabilities. This functionality enables users to:
- Define the report’s selection criteria
- Determine what they would like the report to look like
- Produce PDF, CSV or XML versions of their chosen report
- Pre-load the email addresses of the reports’ recipients, so once created the report can be automatically emailed to the intended recipients
Reports from Velocimetrics can be created intra-day as required or scheduled to run and then sent to the selected email list at pre-set times or if the user wishes submitted for processing by an alternative system. This approach provides users with maximum flexibility.
One-click query execution
Velocimetrics 6 enables users to set-up and save a query with pre-defined query and output format selections. Other users can, subject to having the correct view permissions, simply click on the query definition to run it,
This functionality enables users who may not be overly familiar with the solution to execute pre-defined queries in a single click. In doing so, they can still benefit from the enhanced insight into complex situations that advanced queries can offer, without needing to be an expert system user.
Improved manageability of large volumes of aggregated data for analysis
Whilst Velocimetrics’ aggregators have always effectively handled large data volumes, users were finding that locating figures in deep or broad aggregations could prove time-consuming. To assist users with this functionality, two new aggregation features have been added.
Firstly, users can now access a filter bar to speedily centre the aggregator in their area of interest. So by simply submitting a filter string, possibly with wildcards, the aggregator view will remove all nodes failing to match the filter.
Secondly, you can set up an aggregator so it only retrieves the “top” rows. For example, if the user had an aggregator which was rolling up computed TCP statistics across a very large number of TCP-level conversations, and they were interested in the worst 10 performing conversations related to a certain statistic, Velocimetrics 6 will now effortlessly configure the aggregator to deliver data, ten rows at a time, sorted by the selected statistic.
Through these combined improvements Velocimetrics is able to support an even wider array of business situations, and the business and technical teams Velocimetrics assists benefit from an enhanced user experience.