PlatformOne™ is an extraordinarily flexible software platform for handling real-time data. Its protocol-based architecture means that it can be used for an extremely wide-ranging set of uses.
A system based on PlatformOne™ consists of a series of functional modules connected to each other using a common real-time TCP/IP protocol. Each of these units focuses on a specific task. Here are some examples of the kind of functions that are carried out by some of these modules:
Feed handling
The thing about feeds is that they’re all different. Whoever was in charge of software standardisation took one look at the area of real-time data and decided to give it a miss. So we’ve developed a whole range of feed handlers that understand all these individual protocols and APIs so that they can deliver the data in a common, normalised internal format. Our feed handlers are described in more detail here.
Connectivity
PlatformOne™ can either be used to build a self-contained, standalone system or an adjunct to something else. That something else can either be an industry standard platform such as Thompson Reuters RMDS or an internal client system. We also have modules that send alerts to systems such as ITRS Geneos and SNMP based on the values of data within a PlatformOne™ system. We can even formulate and distribute e-mails using PlatformOne™ data. These are all described in more detail here.
Contributions
We have a long history of providing multi-platform contribution systems, and we believe our offering is one of the most flexible and sophisticated around. There’s more on our suite of products here.
Calculations
PlatformOne™ has a remarkable Calculator module that permits you to apply a range of different calculations to the incoming data. Not only that, but it’s fully extensible: you can write your own plug-in calculation modules using a simple COM interface. There’s more about Calculator here.
Analysis
PlatformOne™ has several modules that can be used for analysing incoming data. Perhaps the most interesting one of these is Tracker, which can monitor the progress of real-time data throughout your network, identifying bottlenecks in real time. Even more impressively, Tracker also generates its own feed of latencies which can then be used to trigger external alerts via SMTP or other monitoring systems. The whole latency monitoring suite is described here.
If it’s data quality you’re interested in, there’s the Data Quality Analyser, which can apply a range of different tests to a data source to determine if, for example, an unusually high value has been identified, or if a record has simply stopped being updated. This is described in more detail here.
For offline analysis, there’s the CSV Logger, which can be set up to capture snapshots of a set of records at pre-specified times and save the results to a CSV file. This is described in more detail here.
Client display
To be honest, we don’t go in for front-end client display modules at JPA; there are plenty of other people who do that sort of thing a lot better than we do. However, we do have a neat Excel add-in in the shape of the LiteStation™ product. This makes it very easy to extract data from a PlatformOne™ system into a spreadsheet in real time. There’s also a counterpart product, called PubStation, that makes it very simple of publish data out of a spreadsheet into a PlatformOne™ system.
Application development
We realise that even with the remarkable array of modules already available off the shelf, there are always going to be new requirements. So that’s why we’ve built a comprehensive set of APIs, which permit application development using Java, COM or .NET.
Simulation
When we were developing PlatformOne™, we realised very early on that what we really needed were some really cool real-time feed simulation tools. So we built some. We think they make a pretty good self-contained simulation environment, too, and they’re described in more detail here.
The PlatformOne™ architecture means that most of these modules can be interrogated in real time to find out their status. Not only that, but the data itself can be interrogated. Each individual record can be asked to provide its application-specific metadata, which can provide information on how a particular value was arrived at. This is one of the most powerful aspects of the entire PlatformOne™ product set, and it means that issues can be tracked down and resolved very quickly, often without having to restart a single part of the system.