The functionalities engineered around the core object store platform not only results in cost reduction but offers many added features like compliance adherence, data search, file sync and share, cloud on RAM, says John Mansfield, President and CTO, Hitachi Platform Division, in an interaction with EC’s Abhishek Raval.
Some of the recent product enhancements done in the private and hybrid cloud space?
There have been many but the major enhancement has been in the object storage space. It’s a new market. We have built a whole suite of products around distributed object store, which is currently, the market share leader in the world.
Elaborate on object storage and how does it score over the traditional file servers ?
We have a large number of customers, whose file servers are spread all over the departments. A large enterprise might have ten to fifteen thousand file servers sharing data in one location. But there is no aggregation. Using the object storage technology, we eliminated the need for the file servers market. The data is now stored centrally and distributed geographically and shared with many organisations. Hence, historically, the cost structure is thirty to fourty percent less and whomsoever is authorised, can have access.
The distributed object storage technology build at the back end is very similar to Amazon. We started in 2007. But we have built functionalities on top of it: A file sync and share; built in security; compliance capability in the object store and now data is shareable amongst anyone in the organisation.
Hitherto, data preservation and simplicity of architecture was absent. The challenges include managing multiple petabytes of data under one management structure; gaining access to that data; ingesting data into it; sharing data within the enterprise and looking for value in that data. A bunch of data intelligence capabilities have been built into it. For example, the enterprise can run analytics, correlation engines, to find every document in the company associated with the name, XYZ. It can indexed to know the relationship XYZ has with other people who are associated with it.
The object store can scale out to tens of petabytes of data. It has been hugely complex. Our major competitor has gone through three or four versions and haven’t got it right yet.
So what’s the USP ?
Above all, It’s the compliance capabilities in the object store system. All the regulatory compliance for healthcare, financial services, government agencies have been catered to.
It’s the separation of the data and the meta data. The national archives of the US uses it for all the administration records. The capabilities that we have been able to engineer over the years enables fast data access after the queries are fired. In being able to find the right data in the pool of petabytes of data. Enabling the indexing and keeping the metadata separate.
The other value proposition is that we have build a portfolio around the object store platform. We have built a cloud on RAM on it. No player in the Industry has been able to build such features. We have designed file sync and share on top of it, natively. This is again an Industry first.
Two of the four largest US banks use your solution. What are they using it for?
These banks use object storage solution for long term data preservation. For one of the banks, almost eighty percent of the data end up in their object storage. Data coming from ATMs, video surveillance cameras, voice calls recordings from call centre, email archives, documents etc. Even the database archives, backup servers are moved into the object store. We can make it referenceable and indexable using the cloud connect feature.
How have some of the Indian organisations benefited from the object store ?
Some of the large deployments in India include Infosys, RBI, mygov.in and the Aadhaar platform.
Infosys was incurring significant costs on backing up their Visa applications, expense status etc. These documents largely remain the same. As a result there was proliferation of the tier 1 storage. After moving to the object store, they were able to solve many problems. Stopping the growth of tier 1 storage. Also, they were able to preserve all the records on the object platform. Thirdly, Infosys eliminated backups and the shipping of tapes from the primary to the DR site for data availability. Thus Infosys was able to drive down the data protection costs by 65%.
Another example is the RBI. Since they are a regulator, all the emails have to be preserved. They are all saved on the Hitachi object storage. Reason: The emails can be preserved, secondly, the emails can be retrieved, leveraging the metadata – by name, date, person, any item. The object store also enables for the forensics.
The entire mygov initiative is on the Hitachi object platform. At any given point in time, two million users are having conversations with the government. The data on the platform can be retrieved and analysed. The government can write applications on top of the platform.
Aadhaar is using the Hitachi’s Pentaho analytics platform to undertake all the analytics done on the Aadhaar.
The cloud service providers are leveraging our object platform, to create new services for their existing customers. There are a couple of CSPs who are using the platform to deliver data protection as a service. They have customers who are primarily using it for colocation and Infrastructure as a service. Henceforth, they want to expand the service to include file sync and share and also to provide preservation as a service.