Items Tagged: SVC
You may recall that three years ago IBM picked up a small Israeli company named Storwize. That company sold an inline compression engine that sat in front of a NAS box and compressed/decompressed file data traversing the network. Its claim to fame was that it was 100% transparent to the application making the NFS call and, while it was indeed inline, it had zero impact on application performance. In fact, since the amount of data sitting on the storage box was approximately 1/3 the original size (average compression factor 3:1), reads were be faster from the disk. And since the Storwize box itself added only marginal latency, while it decompressed the data, the overall performance improved. Up until then, compression engines, which had been around for three decades or more, had always delivered compression at the expense of performance. That was simply the price one paid. Storwize broke the barrier for the first time.
Optimization For Real-Time Storage: IBM’s SAN Volume Controller
Storage virtualization has come a long way in the past seven years. After a false start in 2001, fraught with inflated expectations and product deficiencies, the category fell into infamy. Several vendors disappeared, many others repositioned themselves to focus on the Small Medium Business (SMB) space and yet others reinvented themselves with completely different products. Only one company stayed true to the promise of virtualization from the very beginning: IBM.
IBM’s SAN Volume Controller (SVC) product launched in July, 2003. The company took SVC and nurtured the market, in spite of the fact that many in the market didn’t even want to say the V-word anymore. IBM persisted, fundamentally because the customer could see the potential of storage virtualization and could count on IBM to support them through the early learning cycles.
The payoff for IBM is huge. IBM has now shipped more than 12,000 SVC engines operating behind more than 5000 SVC systems. SVC is a mature, enterprise-proven product that has demonstrated proven ROI to its customers. IBM has shown that SVC and its in-band architecture can scale to handle the largest, most stringent enterprise SAN environments. By doing so, IBM has led the market where others have only slowly followed.
The value of storage virtualization is unquestioned. It helps rein in storage capital expenses (CAPEX) and operational expenses (OPEX) that are otherwise running amok. It provides a forum to perform storage management in a consistent fashion even while the underlying physical storage is heterogeneous and possesses its own idiosyncrasies. In our view, it is also a key building block for the next generation data center that will focus on delivering a variety of services. IBM knew that and held steady. We believe the payoff until now is a shadow of what it is to come, as IBM ties storage virtualization to other efforts, such as server blades and server virtualization.
More recently, IBM has taken one more not inconsequential step in defining what value is when it comes to SVC and virtualization. That step is the introduction of Real-time Compression, integrated into the high performance controllers of SVC as an in-line technology that can be used against production data. In this Product Brief, we’ll take a look at SVC and its historical differentiators, and what compression for real-time, primary storage means for SVC customers – the value is no less than tremendous.