Simba targets big data with new SDK
Simba Technologies Inc. has released a new version of  its SimbaEngine ODBC software development kit, including an updated database query engine that taps into backend database power. 

The Vancouver-based company makes SDKs for developers to create ODBC, JDBC, ADO.NET and OLE DB interfaces. The version 9 of SimbaEngine ODBC SDK adds new customization options as well as allowing for greater optimization, the company says.

The software firm also says SimbaEngine’s Collaborative Query Execution (CQE) function has been extended. The CQE delegates parts of a database query to underlying data stores, leading to faster, more efficient delivery of results, according to Simba CEO Amyn Rajan.
“For example,” he says, “a lot of our partners will come to us and their storage engines have indexing capability. We have, within our SQL engine, the ability to take that query and break it in such a way that we let you use your underlying indexing capability but we’ll do the joining for you.”
Rajan says SimbaEngine is the company’s “secret weapon” that provides developers with a way to link SQL and NoSQL data stores in a database world increasingly dominated by big data. “The question with big data is you’ve got this NoSQL world, and yet you’ve got all these applications that only know how to talk to a SQL backend.”
Randy Hearn, a senior research analyst at Info-Tech Research Group Inc. in London, Ont., says from the looks of it, Simba’s CQE “recognizes the fact that the database engines themselves are quite powerful, whether that’s SQL or Oracle or whatever.”
These engines have the potential to “increase the performance if you allow them to do some of the work,” he says.
And as databases become bigger and more complex, Hearn adds, a tool like SimbaEngine could become ever more useful.
“If I had, let’s say, a very complicated joined SQL statement that went to different databases and potentially had stored procedures running to come up with pricing algorithms… those kind of things are very quickly executed within the SQL engine.”
“What this thing does it it allows the SQL engine to run and execute that stored SQL procedure. And then it will then take all the pieces and put them back together.  But it only requests the result set of the individual pieces.”
Rajan adds that the technology enables developers to harness the query capabilities of many different types of databases. “We have the ability to bolt that onto any backend data store,” he says, “and when I say any data store, we can bolt it onto something that’s like an ISAM, we can bolt it onto something that’s cloud, for example. We’ve used Simba engine to build a driver for We can bolt this onto big data…we’ve used this to build a driver for Hadoop.”
Usually, says Hearn, improving database query times is not a major concern for software developers. “You’re talking faster in terms of fractions and fractions of a second. If something takes 10 milliseconds versus, you know, 20 milliseconds, are you and I as users going to see that? No.”
But it’s when you talk big data that slower query times will become noticeable. “When you start multiplying that by volume then it starts making a difference.”
Hearn large amounts of data coming from different sources could eventually lead to slowdowns in database performance. “Something like this may be the answer to that,” he says.

Related Download
Virtualization: For Victory Over IT Complexity Sponsor: HPE
Virtualization: For Victory Over IT Complexity
Download this white paper to learn how to effectively deploy virtualization and create your own high-performance infrastructures
Register Now