Access Data Faster
Computers process information at the speed of light, it is said. When we're waiting for that report to run or that transaction to complete we often wonder whether that is indeed the case.
Computers can indeed process data quickly, but only if they're instructed to in the right way. And when it comes to servers, they receive so many requests, that even if each request only takes a little longer than it has to, then the server will soon be running a lot more slowly than it could – making you think that maybe you should make that big cost investment in upgraded hardware.
Instead we look at the way data flows throughout your application from end-to-end. We identify where operations are taking too long, and then with your developers, take the necessary steps to fix the problem. Here are couple of examples of where developers go astray.
Asking Questions In The Right Way
Most business data is stored in databases. You retrieve data from a database by “running a query” – asking a question. There are often many ways of asking the question – all of which will return the same answer. Asking the question in some ways may result in an almost instantaneous answer, asking it in other ways will make it appear that the database has hid in a corner and gone into a sulk.
Knowing just how to ask that question is a combination of many factors – knowledge of how a particular database vendor's product works, how your data is stored in the database, the deep knowledge of just how data flows and is processed by a computer – and the experience of writing and re-writing thousands of queries over our years in the industry. So we know how to ask that question – in just the right way.
Imagine going to the local shop to buy a dozen eggs. You could make twelve trips bringing back an egg each time. Or you could send a HGV, fill it with items from the shop, then throw away what you don't need when it gets back. Or you could make one trip and bring back a dozen eggs. Which is the quickest? Which one costs the least?
Software developers often make one of those two errors when accessing data from a database. And such is the complexity of modern software development, it is often not immediately obvious that this is occurring. Moreover, on small datasets often used in testing, performance may be good enough. Then the application goes into a live environment and the application – and maybe your infrastructure – suddenly grinds to a halt.
Catalyst Computing Services are not only experienced with databases, we have extensive experience in writing and maintaining data-based applications, ensuring that data is always accessed appropriately. Moreover, we can mentor and train your own developers so that they can find and fix these problems before their applications cause problems when running live.
Tracing Products Through The Supply Chain
A large manufacturing company required improved reporting of traceability for their products in order to satisfy EU regulation. Each product was built from different components in multiple stages across multiple sites, making the traceability reporting complex.
Data was being captured at the start and end of each production line. On the full scale database this resulted in tens of millions of rows of data. Running a traceability query on this dataset was projected to take in excess of 24 hours.
Using our knowledge and experience of databases we went though a multi-step optimisation process, reducing the time taken to under 15 seconds. This meant that rather than having to submit a query and have the result next day, reports could be generated via a web interface and returned immediately, giving much greater interactivity and far wider usefulness to the business.
So talk to us at Catalyst Computing to find out how we can speed up your flow of information, and give you a competitive edge.