I have to admit that it’s embarrassing when The SSD Guy misses something important in the world of flash storage, but I only recently learned of a paper that Baidu, China’s leading search engine, presented at the ASPLOS conference a year ago. The paper details how Baidu changed the way they use flash to gain significant benefits over their original SSD-based systems.
After having deployed 300,000 standard SSDs over the preceding seven years, Baidu engineers looked for ways to achieve higher performance and more efficient use of the flash they were buying. Their approach was to strip the SSD of all functions that could be better performed by the host server, and to reconfigure the application software and operating system to make the best of flash’s idiosyncrasies.
You can only do this if you have control of both the system hardware and software.
The result was SDF, or “Software-Defined Flash”, a card that Continue reading
IBM today announced its FlashCache Storage Accelerator, a software product that supports flash caching in a broad range of systems. FlashCache operates over three families of IBM servers (System x, BladeCenter and Flex System) and a variety of flash types to accelerate any back-end storage, including non-IBM storage arrays.
Although the cache’s data is dynamically updated to match the random workloads of virtualized systems (i.e. to accelerate VMware), it also improves the performance of Windows and Linux environments.
The cache uses a Write Through policy to solve a number of Continue reading
OCZ has just introduced a new software/hardware combination aimed at accelerating the performance of Microsoft SQL Server. The OCZ ZD-XL SQL Accelerator combines a custom PCIe SSD with caching software optimized for the SQL Server workload. The company boasted that this product won the “Best of Interop 2013” award, a distinction certainly worth crowing about!
How does this product differ from standard caching software? OCZ tells us that the key elements required to Continue reading
- Support for multiple SSDs of mixed sizes and types
- Concatenation of two or more SSDs into a single caching volume
- SSD mirroring for “Safe Write-Back” caching
- A maximum cache size increase from 1TB to 2TB
- 2,048 accelerated objects can be stored in each cache, up from the earlier release’s 255
The company prides itself in the ease with which the software and a server-side SSD can be added to a system without otherwise changing its storage architecture.
With the “Safe Write Back” feature the FlashSoft software has been upgraded Continue reading
At IBM’s Edge 2013 conference last week the company not only extolled the values of flash, as does anyone who has had a flash experience, but it also showed how flash could be made even faster than it already is.
You’re probably already thinking: “Flash is about 1,000 times as fast as HDD – how do you make it even faster?”
The answer is actually pretty simple: compress the data. If there is a limit to how much bandwidth you have going into and out of a piece of storage, you can speed it up if you can reduce the size of the data that consumes that bandwidth.
Of course, that’s not always easy. Compression often slows data access down, even for HDDs, and it could Continue reading
The folks at NVELO recently provided The SSD Guy with some benchmark data comparing their Dataplex software’s performance against the Intel iSRT caching software that is becoming prevalent among Ultrabooks.
For those unaware of these two technologies, they are both caching software that automatically maintains “Hot” data within a low-capacity SSD while leaving “Cold” data on the system HDD. The end result is that the PC performs as if it boasts a large SSD when, in truth, it uses Continue reading
A topic The SSD Guy often brings up in presentations is the fact that SSDs can be used in enterprise applications to reduce server count, a phenomenon often called: “Server Consolidation.” This is a confusing issue, so it bears some explanation.
There are lots of ways to accelerate an I/O-bound application. The most direct one is to speed up the I/O. In the past this has involved some pretty elaborate ways of using HDDs in arrays with striping and short stroking. Many of these arrays cost a half million dollars or more.
Another is to hide the slow I/O speed by Continue reading
SSD maker OCZ has been on something of a tear recently, introducing three new solid state storage products in three weeks:
- Two weeks ago the company introduced the Z-Drive R4 CloudServ PCIe SSD, designed for the data center, in single-card capacities ranging from 300GB-16TB. This product can transfer data at multiple gigabytes per second rates to deliver over a million IOPS.
- Last week saw the introduction of the OCZ-SANRAD VXL enterprise storage accelerator, after OCZ’s January acquisition of SANRAD. This product is flash cache acceleration software for VMware ESX and Citrix Xen virtualized environments that allows Continue reading
Seagate has just announced an upgrade to the company’s Momentus XT hybrid HDD family. Seagate announced in August that the company had already shipped over one million units of its first generation Momentus XT since its May 2010 introduction.
For those unaware of what a hybrid HDD is, the short answer is that it’s a way to get HDD capacities and SSD speeds at a price marginally higher than that of an HDD. Continue reading
One of the SSD Guy’s favorite subjects is caching and SSDs. This is because I wrote a book on processor caches in the early 1990s, and the advent of SSD caches in storage systems hearkens back to the technology detailed in that book.
Caching works well whenever there are two layers in the memory hierarchy since the fast expensive layer can replicate data in the slow inexpensive layer to accelerate the processor’s performance. Continue reading