Jul 24, 2014
In this competitive IT market we are constantly looking for new technologies. Venture capitalists make their fortunes on discovering and promoting new technologies before they cash out and go on to the next big thing. It is tempting to jump on every new promotion to be on the cutting edge. However, many of these technologies may fail or have a very short life span before they are replaced by newer or more comprehensive technology solutions. Unfortunately when this happens with storage, we are often left with the legacy of stranded storage, with tons of data that must be migrated to be usable on legacy, new, and future platforms. The caution when choosing new technology solutions for storage is to ensure that you have a long-term strategy for integrated data management, migration, and retention. With the explosive growth of data, you cannot afford to be locked into storage silos.
Here are two storage technology trends that are popular now, but will fade as soon as consumers discover the silo nature of their implementation and realize the drag that maintenance of silos will have on their storage management costs.
All Flash Arrays: AFA
All flash arrays, AFA, are storage arrays that only contain flash media. The rationale for AFA is that they provide flash performance for all the data that they contain, as well as lower power consumption per capacity than traditional hard disk or hybrid arrays that contain flash and hard disks. The problem with AFA is that they create a data silo, which is not integrated with the rest of the data management pool of storage resources. Data management functions, like data protection, data replication, data security, data privacy, event logging, and archiving, are done separately, if at all. While all the data will get flash performance, most of the data will not require flash performance. In fact most of the data will not require any access at all as the data ages. Flash should be integrated into an enterprise storage platform so that it is part of the overall data management pool, and data can be automatically moved to lower cost, more durable, tiers of storage. Flash and hard disks should not require an either or decision when a hybrid array can provide the best of both worlds. Hybrid arrays like the Hitachi HUS VM and the VSP G1000 can be installed as all flash arrays with better IOPs and throughput performance than AFA, at a comparable TCO and with complete integration into a common pool of data management resources. Since Hitachi did not rebrand the all-flash HUS VM or the VSP G1000, you will not see these products rated in Gartner’s Magic Quadrant for AFA.
My prediction is that the AFA market will be displaced by Hybrid arrays that provide all the benefits of Flash performance integrated into an enterprise storage platform.
Stand alone Enterprise File Synch and Share: EFSS
The trend towards mobile devices as a productivity tool for enterprise knowledge workers has created the need to sync and share files with enterprise security and administration across internal and external desktops and mobile devices. There is a clear need for enterprise file sync and share, but the approach that most vendors have taken is to provide stand-alone solutions, which are not integrated with enterprise data management. This creates silos like the AFA mentioned above. Gartner has published a Magic Quadrant for this market as well (G00261766 July 02, 2014 Analysts: Monaco Basso, Jeffrey Mann, Charles Smulders), but in the first paragraph of their report under Strategic Planning Assumptions they state: “By 2017, less than 10% of today’s EFSS destination vendors will offer stand-alone products, and the rest of the vendors will be absorbed into adjacent markets, such as collaboration, enterprise content management (ECM), mobility and storage.”
This is the approach that Hitachi Data Systems has taken from the very beginning with HCP Anywhere. HCP Anywhere is an EFSS that enables the synching and sharing of files across the desktop and mobile devices wherever the endpoint is. HCP Anywhere is based on the Hitachi Content Platform, which is an object store that was originally developed for compliance with the search capability of an active archive and has evolved into a true, enterprise hybrid cloud storage platform. It can scale to billions of objects, provide safe multi-tenancy, ensure immutability and track the access and change history of objects wherever they reside, on a local LAN, remote branch office, mobile device or in the cloud. Galen Gruman who follows the Mobile market for InfoWorld is quoted in a recent blog post saying: “The Hitachi Content Platform (HCP) announced this week isn’t about mobile data access specifically — and that’s a good thing. Mobile shouldn’t be a separate silo, but another channel to your portfolio of endpoints.”
My prediction is in agreement with Gartner’s take that EFSS will be integrated into the enterprise object storage market.
What will be needed for data growth?
As the amount of data continues to grow from petabytes to exabytes and beyond, we will not be able to manage data if it is stored in silos. Data growth will also require the access and management of new end points and formats. — Think machine to machine and Internet of Things–where we cannot take a piecemeal approach to data endpoints. We need to have a pool of storage resources that spans all the endpoints and common repositories. That pool will need to be virtualized, so that it can integrate new technologies and manage itself automatically based on policies or business events. I am not exactly sure how that will all play out except to avoid niche storage solutions that create data silos.