Archive | Cloud Computing

I Can’t ‘Contain’ Myself


I blogged about this topic last week, but looking forward toward all of what will be announced at VMworld next week has made me want to revisit the subject again and make some predictions on what the show will be all about this year. I literally couldn’t (pardon the pun) ‘contain’ myself 😉 I do quite a bit of thinking around this topic every year so as to inform our Marketing team on what our messaging might look like as we approach VMworld, so I’ve been thinking about these topics for some time now…

What exactly do containers mean for the virtualization admin, storage admin or otherwise? 

Containers such as Docker and Rocket, among newer generation technologies, aim to provide abstraction much like virtual machines, but without the added effort of virtualizing an entire OS and all of the supporting drivers. A much more efficient method for application delivery, which I have blogged about before and have had conversations with colleagues, such as Scott Drummonds, about in the recent past couple of years. Around the time that CloudFoundry was first announced, we talked about what application delivery would look like without the need for an OS. Obviously it would make for a higher performance, transient, web-scale system that would deliver the application without all of the fluff required by a typical virtual machine. We didn’t know when it would happen, but definitely glimpsed the possibilities and implications in IT once this dream came true. We are starting to see the results of that now, with “containerization”. VMware is at the forefront of this innovation (along with Docker and CoreOS) with projects such as Lightwave and Photon. You will obviously see these heavily featured at the show, along with other DevOps and Microservices related topics. I see these trends as a 2nd or 3rd iteration of the cloud native apps paradigm that CloudFoundry hinted toward a few years back. Paul Maritz talked quite extensively about this during his tenure at VMware, stating something along the lines of: now that customers have realized the economic benefits of virtualization, they will start re-writing their apps for a cloud native future. This didn’t happen right away, but I believe that it’s telling that we are now seeing the results of this effort. This definitely follows with VMware’s “Ready for Any” theme for this year, I think you’ll agree.

Storage and the next wave of containerization…

As an emerging, cutting edge storage provider, we at Coho Data are also innovating around this trend as it relates to storage, especially persistent storage, which is in its nascent phase. We will not only be demoing some microservices running directly on our array, but we’ll also be talking about what the microservices future will look like in the context of storage. Think data-centric, not workload-centric, VM-centric or otherwise. We care about storing and managing data regardless of the abstraction it sits in.

This is just a taste of what we have in store. Come visit the Coho Data booth (#1713) or attend our CTO, Andy Warfield’s breakout session on the topic to learn more about what we have in-store.

Looking forward to seeing you there!

10,263 total views, 6 views today


Hybrid or All-flash?

spinning-flash-hybridI’ve been seeing a lot of commentary of late (actually it’s been happening for a while, but now that we’re approaching VMworld, I guess I’m more conscious of it) about the value provided by hybrid storage systems from a cost perspective, while at the same time, the all-flash vendors out there are touting the storage efficiency features (they have to!) of their platforms, to get to a price point that run-of-the-mill customers can stomach, while providing very high performance.

There’s been lots of FUD about how the economics of hybrid systems will always be more cost effective when compared to all-flash systems. By contrast, the all-flash vendors will say that now, with the announcement of a flash device that is larger than the biggest hard disk, that all-flash will win out, once prices come down. Finally, a select few vendors are also saying that because they provide both hybrid and flash systems that they are (falsely) offering customer choice. I beg to differ.

How about a third option all-together, or if you will, no option at all.

What exactly do I mean by this???

Well, since the announcement of our 2.x release, Coho has been able to host both hybrid AND all-flash nodes within a single cluster, in a single namespace. Not only is this easier for the customer to set-up and manage, but a much better story from an economic perspective as well.

Consider the fact that, because we leverage software-defined networking together with software-defined storage, that we can firstly, start with a single 2-node cluster, unlike any of our competitors in scale-out, that must start with 4 or more. Second, the fact that we can place data in the appropriate tier of storage based on a per-workload view (we present as NFS today) of the flash utilization over time, and we make even more efficient use of the performance and capacity in the system, while simplifying it’s deployment for the customer. This means that if our analytical data indicates that you are running out of high performance flash, you can expand your system with an all-flash node OR if the analytical data says that you’ll run out of capacity while still maintaining the same level of performance, you can expand with a hybrid system. We take a data-centric view of the customers’ workloads, so that they are freed from making these choices; and we’ve found that the message of simplicity is resonating quite well, thank you!

Expand this to next generation flash media types such as NVDIMM, PCM, RRAM, MRAM, etc. and you can imagine why having MANY different types of flash memory in a single cluster, single namespace will become a must have feature.

Coho Data is on the path to deliver this today. Our platform was built with this in mind from the word “Go”.

It’s time to upgrade your thinking!


11,393 total views, 5 views today


Web-scale Economics… and Innovation?!


What is Web-scale?

A good percentage of those of us out there in the trenches of enterprise IT have probably heard the term “Web-scale” thrown around. It, like many IT terms, is equal parts marketing term and technical term, hence, not-so-well-defined… and as a result, open for interpretation. My take on Web-scale is that it’s, first and foremost, a way to architect IT systems for enterprise, incorporating the best elements of public clouds. While it is very hard to mimic the architectural scale and resiliency of public and private clouds from AWS, Google, Facebook and others, one can easily see the benefits of distributed, shared-nothing architectures, API-driven automation and orchestration, self-healing application stacks… and in Coho‘s case, closer integration of the network with the storage.

The Coho approach to Web-scale has some unique elements that separate us from the other vendors that purport to do it. Hyperconverged vendors are for the most part confined to growing all datacenter resources simultaneously. Scaling all datacenter resources at the same time doesn’t necessarily make sense, unless your environment has very uniform workloads. My guess is that if you are a typical small/medium or enterprise, your compute, network and storage requirements don’t scale at an identical rate, thus performance gets left on the table, or you end up licensing software that you don’t need in order to grow your footprint. With Coho, we allow the customer to scale the compute independent of the network and storage. As you add building blocks to a Coho scale-out cluster, you add 40Gbps (or more) of network bandwidth along with multiple TBs of PCIe NVMe flash. This is a hard requirement if you expect the cluster to exhibit linear performance scaling as you add capacity. Adding flash without the adequate network bandwidth to push the bits over the wire is a lost war before the battle even begins!

This brings us to the economics part of the discussion as it relates to Web-scale…

Converged (non-hyperconverged) systems that incorporate increased network capacity along with the storage, such as Coho, give customers the ability to incorporate the best elements of public clouds with the security and performance that can only be achieved with on-premises infrastructure. This simple fact has afforded us an opportunity to talk to customers in the terms of $/GB/mo that they are likely to see quoted from Amazon and others. The shift toward OPEX pricing is already top of mind for a great many CIOs, so it serves as a convenient reference point for us when we talk with customers. Even with operational costs figured into the economics, we often talk about prices that are 1/2 to 1/3 the cost of AWS. Now let’s put a qualifier here… we’re not talking Amazon Glacier or the cheapest of the cheap that Amazon offers, but rather AWS EFS (Elastic File System) service which is advertised at around $.30/GB/mo, all-the-while preserving the jobs of the internal IT teams, and preserving corporate IP (intellectual property) security and providing better performance! Don’t even get me started on the costs associated with getting data into/out of AWS once it’s in their cloud. You ever heard of data gravity?

But wait, there’s more…

Since Coho is innovating by creating unique storage services directly on the array, by leveraging Docker, Kubernetes, VXLAN and other cutting edge technologies, we are able to offer alternatives to AWS, without the need to move to the public cloud. This is the move toward “microservices” that you may have heard about. As a matter of fact, not only will Coho be demoing these technologies, in the form of on-the-fly transcoding, a search appliance and more, but our CTO, Andy Warfield will also present a breakout session discussing this very topic. Why bother going to AWS for services that you can get as free upgrades with a paid support contract?

In my opinion, Coho is not only at the forefront of what Web-scale was intended to deliver, but taking it to a whole new level. Look for us at VMworld (booth 1713) to find out more… we’re looking forward to talking with you!

9,061 total views, 3 views today


“Pets” vs. “Cattle”… In the Context of Storage?


By lamoney ( [CC-BY-SA-2.0 (], via Wikimedia Commons

I’ve been thinking a little bit (more than usual) lately about the crossroads we are at in the IT industry today. I’ve been reflecting back to some early posts that I shared way back when virtualization was the tech de rigueur. Not only that but the fact that my current company Coho Data is at the nexus of this crossroads, if you will. Since we talk “web-scale”, “scale-out” and a multitude of other buzzwords in today’s IT world, it’s interesting to explore some of those in the context of cloud and distributed systems that form the new reality of enterprise IT computing.

When dealing with cloud computing proximally or otherwise, it’s likely that you fall within either the VMware camp or the OpenStack camp (or both) today. Some would say these solutions are at opposite ends of the cloud software spectrum. You may also have heard the term: “Pets vs. Cattle” in reference to your servers, i.e. a Pet has a name, requires constant patching, updating and altogether expensive maintenance… whereas Cattle are nameless, can be removed from the system and replaced with new gear and be online again doing their job without skipping a beat.

Well, what if it were possible to have a zoo and a farm all-in-one? and what about for storage?!

Normally when you think of storage, its persistent nature requires it to be a Pet and not Cattle, but with today’s more modern storage architectures, I’d like to propose that this isn’t necessarily the case. You can have both persistence of data and statelessness of the underlying components at the same time. Bear with me for a minute while I reason through this…

With a scale-out, shared-nothing node architecture, you have the ability to add and remove nodes on the fly without worrying about the health of your data. As you scale to larger number of nodes, you care about each node even less. Despite the fact that you have a greater quantity of data in the system, the importance of any one individual storage node is reduced. Add to that the fact that a well-built self-healing, auto-scaling system can heal itself faster when there are more “cattle” on the farm.

As a function of this architecture you can also remove nodes in much the same way, allowing you to return leased equipment or installing newer, more dense and performant nodes into the system with everything working in a heterogenous fashion, and without skipping a beat. This is great from a TCO perspective as well.  It’s much better than being locked in with a fixed amount of high performance flash and capacity spinning disk for the next 3-5yr spending cycle. Extend this even one step further and you can imagine being able to automatically order new hardware to expand the system, adding it, then shipping back the old to the leasing company in a regular, predictable fashion.

One element of cloud scale systems that allows this to happen is extensibility, being able to easily extend a system beyond it’s original reason for being. Typically this is enabled via APIs and all of today’s next generation storage systems are build from the ground up to support this type of integration. Being able to organically adjust to customers’ needs quickly by offering APIs, toolkits and frameworks, is a key ingredient in delivering web-scale!

The interesting part of this whole discussion is that despite the importance of persistence in the storage world, given the right architecture we CAN indeed have the best of both worlds. Look at Coho’s scale-out enterprise storage architecture and you can see that we very much have a combination of the elements of both Pets and Cattle. We support the best from either architecture as well as any modern storage system should

Here are some examples:

  • Pets like NIC bonding for high availability – we’re cool with that
  • Pets like to be managed carefully and thoughtfully – we build intelligence into our storage, but also give visibility to the admin
  • Cattle can be auto-scaled by just plugging in a new node and allowing the system to grow – we do this as well
  • Cattle are designed to accommodate failures – we build our failure domains across physical boundaries so that their is no single point of failure
  • Pets like to have constant uptime – refer to previous feature of cattle above; accommodating failure means the system stays online if a component fails
  • Pets like to have high availability – we do this as well, allocating a minimum of 2 physical nodes in a single but shared nothing hardware design
  • Cattle work only when there is shared nothing architecture – utilizing independent nodes with object-based storage allows us to provide this as well!

At Coho we see the need for these differing approaches to computing from a storage perspective. We started out providing storage for VMware workloads and customers seem to like how we’re delivering on that so far. In addition, we see the need to support OpenStack from a storage perspective as well and are currently offering a tech preview of our OpenStack support. As a matter of fact, if you’re interesting in becoming a BETA participant for OpenStack, you should definitely get in contact with us.

Thanks for reading!

6,354 total views, 3 views today


OpenStack Summit 2014


Tonight I’ll be heading to Atlanta for OpenStack Summit next week. I’ll be representing Coho Data and this will be my first time attending a conference not directly related to VMware (or a partner of VMware). I am very much looking forward to some good conversations around what users are expecting in a storage product that supports and enhances the use of OpenStack. I’ll be attending some sessions to get up-to-speed but will be spending the rest of the time at the Coho booth. I’m sure I’ll run into lots of familiar faces both new and old. Stop by and introduce yourself, or maybe we’ll see each other at one of the various events going on outside the official venue.

See you next week!

7,176 total views, 5 views today


Powered by WordPress. Designed by WooThemes