The RamSan-20 SSD weighs in at 450 GB, and it’s based on single-level cell (SLC) NAND type flash memory. It plugs into a PCI-Express x4 slot on the server and has a 333 MHz PowerPC chip as is main controller. With its banks of flash memory modules, the RamSan-20 can support 120,000 sustained I/O operations per second (IOPS) on random reads and 50,000 IOPS on random writes. With a 70-30 per cent mix of reads and writes (which is typical for a lot of workloads), you get about 80,000 IOPS. The whole unit burns about 15 watts, which is about what a stick and a half of DDR2 main memory inside a server does. The unit has error correction electronics as well as four RAID 5 chipsets (made by Xilinx) to make sure the data stored on the SSDs stays clean.

The RamSan-20 PCI-based SSD will be available early in the second quarter, according to Texas Memory Systems. The unit will sell for $18,000. (Good business, if you can get it).

Texas Memory Systems also sells lines of storage arrays based on flash and main memory. These plug into servers just like disk arrays, but they run a hell of a lot faster and cost a hell of a lot more money. ®

Medium-sized organizations are increasingly being asked to improve information management to more efficiently support their businesses while keeping costs down. Storage technology providers, in turn, have traditionally responded with midrange‘ storage systems, frequently defined as scaled-down and compromised versions of enterprise class arrays. Medium-sized businesses, however, are in need of ―enterprise-class‘‘ information management and infrastructure technologies—without the enterprise price tag.HDS is once again delivering innovation—the level of capability and scalability in the Hitachi AMS 2000 family is compelling and extremely valuable to end-users.

Setting the Record Straight

Data Domain has increased the speed of its deduplication to between 50 and 100 percent through a software update.

The speed boost varies from 50 per cent on DD510, 530, and 580 models, 58 per cent on the DD565, up to 90 per cent on the DD690g and DDX array, and 100 per cent on the DD120.

The DD690g had a 1.4TB/hour throughput rating when it was introduced in May last year. Now it is 2.7TB/hour (with Symantec NetBackup OpenStorage and a 10GbitE connection).

It makes you wonder whether the code was that bad; clearly there was room for improvement. What Data Domain has done is to get its deduplication code executing better on multi-core CPUs so that more is done in parallel. The code, using a technology Data Domain calls Stream Informed Segment Layout (SISL), has been tuned so that it makes better use of the available cores.

Data Domain’s platform operating software has revved from DD OS v4.5 to 4.6. Shane Jackson, senior director for product and channel marketing at Data Domain, contrasts competing vendors who, he says, rely on adding disk spindles to boost deduplication speed, with Data Domain’s reliance on CPU speed. “Intel shows up with a faster processor more often than Seagate shows up with a faster drive,” he says.

That’s over-egging the pudding, as all dedupe vendors rely on software and disks. Data Domain happens to have, it appears, as good as if not better software algorithms than most, certainly enough for it to suggest its products can be used to deduplicate some primary storage applications.

The neat aspect of this is that it’s widely expected to introduce new, Nehelem-boosted hardware later this year. With 8 cores available there should be another doubling or near-doubling of performance compared to the current quad core Xeons being used. That means a DD690-type product could ramp its performance up to 5.4TB/hour, meaning 90GB/min or 1.5GB/sec.

Sepaton and Diligent, now owned by IBM, emphasise their deduping speed and NetApp also pushes its ASIS dedupe into some primary data deduplication applications.

Two thoughts: first, it looks as if a deduping race is on. Secondly, it begins to look as if inline deduplication is quite viable for the majority of backup applications.

Offline dedupe vendors say that, to keep backup speeds high you really should land the backup data uninterrupted by any processing and dedupe it afterwards. At speeds of up to 750MB/sec now and with 1.5GB/sec speeds coming, Data Domain would say that most backup applications could be deduped inline and avoid the need for a substantial chunk of disk capacity kept aside to land the raw data.

Businesses today are focused on surviving the current downturn in the economy. Everyone seems to be hedging their bets, hoarding their cash, trimming the fat, and sweating their assets. Although governments have pumped massive amounts of cash into the banking systems, credit is still tight. Banks are reporting increases in deposits as people reduce spending and cash out of stocks or other investments to preserve what they have left.

Most data centers are sitting on a gold mine of under utilized storage capacity. With utilization around 30%, data centers can defer purchase of storage capacity by increasing the utilization of existing storage to meet increasing data demands. Technologies like storage virtualization to consolidate IT assets, dynamic provisioning to increase utilizaton and reduce operational costs, de-duplication and other technologies to reduce data storage requirements, and active archives to remove stale data from the working set, will help companies survive during this down turn. 

At some point there will be a recovery from this current down turn. My guess is that this recovery will be as rapid as the fall and there will be a sudden demand for capacity. The money that has been sitting on the sidelines will be rapidly converted into goods and services to meet the pent up demand of consumers, once confidence has been restored.

This can be a double whammy if IT has not prepared for this type of sudden recovery. After paring down their staff and squeezing the most out of their aging IT assets, they may not have the capability to meet this surge in demand and be competitive. If they have not implemented tools like virtualization, dynamic provisioning, de-duplication, and active archives, the same tools that are required to survive the down turn, they will not have the agility to respond and be competitive.

When the recovery comes, the playing field will have changed. Storage assets will have aged and capacity utilization will be reaching its limit. IT will no longer have the luxury of taking 6 months to a year to implement a new storage frame with 100’s of TB of new capacity. Migrating applications to new technology platforms will have to be done without any down time. Experienced operations, applications, systems, and development people who can ramp up quickly will be in high demand. Costs will be inflated and time to market will determine who will be the winners and the losers.

Some will survive this down turn only to be left behind in the recovery. Now is the time to plan for the recovery while demand is low. Don’t be trapped into buying the lowest price capacity and ignore the operational benefits of virtualization and dynamic provisioning. Do an assessment of where you are in terms of operational as well as capital costs and set a strategy for where you need to be, not only to survive but to prosper with the recovery. Do a storage economics assessment and set a benchmark that you can use to show the value of improvements and get credit for it. So many times we add value but we are not able to quantify it to the financial people because we never set a bench mark in the beginning. If you have the right tools and strategy in place to survive the downturn, you will be ready to survive the recovery.

Multi-dimensional scale has already appeared on the market and it is a core requirement of this new generation of file-based storage architectures. Scale-out, the ability to independently scale and tune bandwidth, processing, and storage capacity on the fly—all while managing the file system and single global namespace—will be the new backbone of file-based storage solutions. Scale-out storage architectures are significantly different than the monolithic, scale-up storage architectures (e.g. traditional NAS or SAN systems) that developed in the Distributed Computing Era. Scale-out has been around for a while, but it has been tucked away in a corner, mostly used in HPC and scientific computing environments, as well as media and entertainment. But the advent of new computing use models such as Web 2.0, SaaS, and SOA introduces the requirement for scale-out in commercial enterprises.