Skip to main content

Storage Virtualization: Gear6 VP interviewed;CacheFx's Centralized Caching Mechanism discussed

This was an truly interesting briefing i have had in the storage. Gary was discussing about the most neglected part in the storage. Scalability by performance! Ok so if we breakup the vendors we have today:

  • Traditional model: add bulks of disks and expect the disk scalability problems to go away. this is old and very cost-prohibitive.
  • Smarter vendors: Pack all kinds of disk in a box and sell disk per disk on-demand model. This is cheaper but does by no means address your problems in performance and caching scalability.
  • Niche vendors: Should start addressing the performance problems as a default built-in option in their torage delivery model. This helps them grow painlessly.
Anyways check out our interview here:

Tell us about yourself, why and how you got started at Gear6?

I’ve spent the last decade in the network storage industry working at several systems and networking companies. One thing that always surprised me was the type of configuration acrobatics that IT administrators often had to go through in order to configure storage for performance. Typical exercises include short stroking drives by only placing data on portions of the platter, configuring extra large stripe sizes that require very long rebuild times in case of drive failure, and general disk over-provisioning.

When I heard about the Gear6 approach to remove this pain from the daily activities of storage and data center administrators, I knew the company was on to something and came on board.

Who is Gear6 and why yet another NAS firm?

Gear6 has a clear mission to accelerate data center I/O so that customers boost application performance and reduce total storage costs. But unlike other NAS companies, our approach is quite different. Gear6 enhances existing storage infrastructure with scalable caching appliances that retain current file systems and management procedures. We don’t believe that the world needs another file system, or even another storage system for that matter. What existing storage systems often need is a performance boost to keep up with ever-increasing I/O workloads. So we’ve tailored our products to add incremental technology without doubling-up on existing solutions that are already deployed.

Can tell us about the I/O negligence against just adding capacity?

Storage systems really need to be judged on three performance-related metrics: I/O operations per second (IOPS), latency, and throughput. Historically, throughput has been the main measure promoted and considered by users which only tells part of the story.

Disks are useful in achieving throughput, and if you add enough disks, throughput will inevitably increase. However, because the access time of disks is limited by the rotating mechanical media, latency and IOPS do not improve by adding more disks. Specifically, latency does not necessarily go down, and IOPS do not necessarily go up.

So effective storage deployment requires attention to all three metrics.

What is CACHEfx technology and how does it work?

CACHEfx is the name of our product line of scalable caching appliances. These high-speed, high-capacity memory based appliances attach to the network to provide caching for any number of storage systems to any number of clients. The caching appliances enhance the existing storage systems by offloading intense I/O activity to the cache where requests can be served from memory instead of disk. This typically delivers an immediate performance increase of 10-50X at the I/O level.

Our Reflex OS software makes use of sophisticated clustering technology so that the cache capacity appears as a single coherent pool. The clustering capabilities also allow for dramatic performance benefits since each cache module within the appliance can operate independently and simultaneously, resulting in a massive fan out of I/O performance. Finally, we selectively apply a set of cache services that optimize caching profiles for particular applications to get the most effectiveness and efficiency from the cache pool.

How is your caching superior to traditional device caching?

Caching is a tried and trusted technology that has been used throughout data centers as long as they have been around. We have caching at the CPU level, the motherboard/server level, within storage systems, and all the way down to the disk drive. Those implementations will stick around and continue to help improve overall system performance.

However, many data centers are facing increasing workloads that are largely impacted by demanding I/O requirements. Simply adding cache to individual end-points such as a server or storage system can, by its very nature, only improves performance at that single system.

As customers recognize the need to make increasing investments in caching, the benefits of placing that resource in the network become more apparent. Now the cache can apply to any storage system and any application server. Plus, that caching resource can scale independently of disk-capacity. These are similar benefits to what we saw years ago with the migration from direct-attached to network storage. When the loads get intense, it makes sense to separate certain functions and enable them within the network.

Tell us a bit about your product line?

The CACHEfx product line is a simple, scalable range of appliances that are suited to a variety of environments. Each product is packaged as a clustered appliance with multiple cache modules. The product line comes in sizes that range by IOPS, throughput, and cache capacity. Low latency is a Gear6 hallmark that remains in the [½ - ??] millisecond range across the products.

about a key advantage of the Gear6 product line is that each model can easily scale. So if customers find that they are getting a big boost from the current product and want to increase their workload even further, they can add to the caching appliance on-the-fly. This is quite different from how customers would have to configure or reconfigure storage systems for performance. Essentially we provide a “big dial” to tune up IOPS..

How is your pricing model?

As a product company, we sell each appliance and the corresponding software as a unit, and then offer a maintenance and support contract for that product. Our new G100 model starts at just over $100,000. Other products scale up from there.

If I were involved in a decision making process with a customer, where can I introduce Gear6? What kind of client should they be?

The common characteristics across our customers are a need for performance coupled with a desire to reduce total systems costs and improve manageability. Typical symptoms of storage bottlenecks include application brownouts and over-provisioning disks for performance. We posted a short video on our company blog about the Top 10 Symptoms of Storage Bottlenecks which can be found here:

What about competition? Do you have any?

There is always competition, but the thing we see the most is IT administrators who continue to over-provision disks because that has been the conventional wisdom. And while that may have been a cost-effective method in the past, today’s ongoing costs of power, space, and cooling make any over-provisioning method extremely painful.

We expect others to follow our network-based caching approach, but Gear6 is leading this market and allowing customers to change the way they buy and configure storage.

Do you complement virtualization or offer a new approach to application delivery altogether?

We see server virtualization as yet another driving force behind increased I/O loads and Gear6 products as a perfect way to enhance server virtualization deployments. Before server virtualization, it was relatively easy to segment and isolate I/O loads because of the physical separation of machines. With server virtualization, a single server might have 5 or 10 virtual machines which don’t necessarily communicate with each other regarding the use of their I/O. So a dense VM environment can often lead to chaotic I/O patterns. This is a perfect opportunity to enhance existing storage solutions with scalable caching appliances.

Will you also explore other storage technologies?

Our appliance approach delivers best-in-class, industry standard hardware with Gear6 software. This strategy allows us to make use of all new technologies that appear within the x86 ecosystem. So absolutely, we have and will continue to explore other storage technologies within our appliance to suit market needs.

What are your expansion plans in EMEA and APAC region?

We are enthusiastic about the response we’ve received from customers in both EMEA and APAC. Those are areas we plan to expand in with the help of qualified partners. To date, our business primarily focuses in North America; that said I/O bottlenecks are not geography specific, and we foresee a substantial, global market opportunity.


Popular posts from this blog

Security: VMware Workstation 6 vulnerability

vulnerable software: VMware Workstation 6.0 for Windows, possible some other VMware products as well type of vulnerability: DoS, potential privilege escalation I found a vulnerability in VMware Workstation 6.0 which allows an unprivileged user in the host OS to crash the system and potentially run arbitrary code with kernel privileges. The issue is in the vmstor-60 driver, which is supposed to mount VMware images within the host OS. When sending the IOCTL code FsSetVoleInformation with subcode FsSetFileInformation with a large buffer and underreporting its size to at max 1024 bytes, it will underrun and potentially execute arbitrary code. Security focus

Cloud Security: Eliminate humans from the "Information Supply Chain on the Web"

My upcoming article, part - 3 data center predictions for 2009, has a slideshot talking about the transition from the current age to the cloud computing age to eventually the ideation age- the age where you will have clouds that will emote but they will have no internal employees. Biggest management disasters occur because internal folks are making a mess of the playground. Om's blog is carrying an article about Cloud security and it is rather direct but also makes a lot of sense: I don’t believe that clouds themselves will cause the security breaches and data theft they anticipate; in many ways, clouds will result in better security. Here’s why: Fewer humans – Most computer breaches are the result of human error ; only 20-40 percent stem from technical malfunctions. Cloud operators that want to be profitable take humans out of the loop whenever possible. Better tools – Clouds can afford high-end data protection and security monitoring tools, as well as the experts to run t