Every year we highlight 10 companies and technologies to watch. Our selection is driven primarily by the fact that the companies we mention have briefed us in the past 12 months, and hence we are reasonably familiar with their capabilities. Naturally, we tend to choose companies and technologies that are distinctive, innovative and align with what we consider to be the direction of the IT industry.
Here is this year’s list, arranged in alphabetical order to avoid any suggestion that ranking is involved:
- Automic: Automic does business automation from the ground up. The company, once known by the amorphous name of UC4 Software, rebranded in 2013 and has an integrated a portfolio of what I tend to think of as orchestration software. It schedules the execution of applications at every level (from trivial scripts to mission critical software) and in every direction (between different data stores and platforms within the data center and between data centers, involving both private and public clouds). My thought is that as everything becomes more and more virtual and physically distributed, we’ll need software that can hold it all together – managing and monitoring the execution of all the business’s applications as if they were running on a single server. That’s what Automic does, and it does it well. We expect it to prosper accordingly.
- BigML: The ML, of course, stands for machine learning. We were bound to encounter some interesting machine learning products last year, as the concept gradually acquired a more magical and promising glow with every passing month. Because the sheer muscle required to exploit machine learning algorithms is now both available (via the private or public cloud) and inexpensive, a number of vendors are vying to enable the use of this technology in various ways. BigML, to its credit, has concentrated on simplicity and drag-and-drop ease of use. They offer a versatile service with plans and modes of operation to suit everyone from the individual researcher to the multinational corporation. If you’re looking at predictive analytics, you might like to take a look at BigML.
- CloudPhysics: A natural consequence of big data is that some software companies will develop an analytics capability that targets a specific and deserving area of application. This is what CloudPhysics has done with big data analytics. The application is the management of virtualized infrastructures. The software continually collects operational data from your virtualized infrastructures and analyzes it to work out how you can use the resources most effectively. Recommendations are generated dynamically and hence can be applied immediately. Basically, you turn it on, and it helps you manage resources and resolve performance issues from the get go. It is, of course, a SaaS offering.
- EXASOL: This year we could have dedicated a post to the 10 database products to watch in 2015. There has been a stunning amount of innovation in the database world. EXASOL is one of the products we’re including, partly because it claims to be the fastest database on the planet – and it may well be, given that it holds the TPC-H benchmark record. It combines in-memory technology, columnar compression and storage, and massively parallel processing to produce its blistering speed. It has modules for OLAP cubes, SAP data integration and Hadoop integration via MapReduce. Like most newer databases, it has a cloud offering, but it can also be bought as an appliance. It targets analytics applications, naturally.
- FoundationDB: FoundationDB is the product I thought nobody would ever build. It advertises itself as a multi-model database, which means it can present, for example, as a SQL database, or a graph database or a document database or….well, to be honest, that’s the point. When it comes to database engines, what you really want is a database that is ACID compliant and offers good performance. Such things are qualities of the database engine underneath, which gets the data and puts it away as ordered. SQL, graph and document queries are all about how the developer wants to receive the data when it is delivered. So why not build a database which separates the logical view of data from its physical storage in such a way that the developer neither knows nor cares how the data is stored, but can view it according to whatever data model makes sense for the application he is building? That’s what FoundationDB (the company) has done, and that’s what FoundationDB (the database) is. Take a look.
- Saffron Technologies: IBM may have invented the term “cognitive computing,” but it isn’t the only player in that space. Saffron Technology has a software platform that can genuinely claim to be cognitive; it applies reasoning to the field of information it is pointed at, which could be just about anything you please. The platform ingests data from disparate sources and then digests it, incrementally learning from new data and adapting to a changing picture in real time. It connects the dots and thus creates knowledge. Businesses use Saffron in diverse ways: anticipating market trends, optimizing processes or mitigating risks.
- TimeXtender: We have long been fans of WhereScape, a data warehouse automation company. TimeXtender offers the same kind of capability: modeling, building and maintaining data warehouses. You might expect, then, that the two companies would be toe-to-toe in the market, desperately fighting it out. However, this is not at all the case, simply because they target different database platforms. TimeXtender focuses entirely on Microsoft SQL Server, and as far as we can tell, it has little competition. In fact its main competition is the in-house SQL Server team, and they don’t quite explore data warehouse automation. As you would expect, TimeXtender’s TX2014 product works seamlessly with Visual Studio and all the associated Microsoft tools. The bottom line is that it dramatically reduces the time to build a data warehouse.
- Tokutek: Tokutek is a very unusual company. It was formed to exploit Fractal Tree indexing, an indexing method which was developed at MIT and which is genuinely better than the B-tree. How much better? It’s hard to say exactly, but Tokutek offers 50x the performance of the typical B-tree while reducing database or file size by about 90%. Given that the B-tree is a feature of almost all databases, you might have thought that Tokutek would have built itself a sparkling new database with its lightning-fast indexing system. But it chose not to. Instead it built Fractal Tree add-ons to different databases. It currently supports MySQL, MariaDB (via TokuDB) and MongoDB (via TokuMX). Tokutek products are also open source, with the company focusing mainly on services and support.
- Voltage Security: I came across Voltage Security quite a while ago. The company had unique security technology in the area of identity-based encryption. Nowadays they also have format preserving encryption, secure stateless tokenization and stateless key management. These terms may mean nothing to you – unless you are deep into security technology there is no reason why they should – but the point about them is that they are efficient and important security capabilities. Often, the problem with IT security is that it gets in the way. Voltage Security’s products do their best to get out of the way. And of course, that has allowed Voltage to make its technology highly scalable for the world of big data.
- WebAction: We figured we had to include a real-time development environment in this year’s list, so we chose WebAction, a fairly new entrant into the market, but one that has built a coherent and easy to use platform primarily for developing real-time analytics applications. WebAction focuses on a marriage between development speed and speed of implementation. In theory, tools of this kind have potential applications everywhere, but in reality, the sweet spots are computer security, retail, data center management and risk management – applications where many disparate data sources may need to be streamed and analyzed in flight in order to respond proactively to what is happening right now.
So that’s it for this year, though due to the high emergence of impressive database products, we should include some honorable mentions: Citus Data, JethroData, ParStream, RedPoint Global, SpaceCurve, SPARQL City, Splice Machine, Stardog – and that is not an exhaustive list. Some of the companies we’ve identified will inevitably get attention in our 2015 research program, Making Sense of Hadoop.