Clustered, Scale-Out Storage
A CEO Interview with PeakColo
Agile Cloud Storage Is Critical to IaaS Success
Delivering an always-on cloud infrastructure requires a storage platform that delivers scalability, automation and agility. For a company like PeakColo, the stakes are huge: The success of its cloud services depends upon the performance of the underlying cloud storage platform.
PeakColo, headquartered in Denver, is an infrastructure-as-a-service (IaaS) provider that builds cloud solutions for channel, value-added reseller (VAR) and managed services providers. PeakColo’s channel partners tailor these “white label” IaaS solutions to meet the specific needs of their end-user customers. PeakColo has selected NetApp® storage solutions to deliver its best-of-breed cloud services.
Here’s what PeakColo CEO Luke Norris had to say about the cloud storage platform that supports his rapidly growing company.
NetApp: Can you provide a brief overview of the cloud services PeakColo offers?
PeakColo: We view infrastructure-as-a-service as the foundation of the cloud computing ecosystem. We deliver the infrastructure and the opportunity for software-as-a-service and platform-as-a-service companies to build on top of us.
We’ve distributed our cloud nodes in major cities throughout the continental United States, allowing us to drive down latencies for the end users of our services. These very low latencies enable our VARs to offer production applications such as Exchange, Oracle and SAP via a cloud service, as well as more traditional cloud offerings such as backup, disaster recovery and virtual infrastructure.
NetApp: We often hear that standardization and automation are critical to the success of a cloud project. What does your storage infrastructure look like today, and how is it managed?
PeakColo: Our storage infrastructure is entirely built on NetApp Cluster-Mode technology. It’s a very large, multi-petabyte system that uses many FAS3200-class controller heads. With other cloud storage providers, we’d have to manage this environment as multiple discrete systems. The Cluster-Mode technology allows us to manage it as a single system and enables us to keep our staffing at a good economic level. Our management efficiency flows back to our customers as cost savings, even as we experience rapid capacity growth.
Automation of the environment is an imperative. The ability to make heavy use of the command-line interface with Cluster-Mode was extremely useful to us. We’ve been able to combine scripting with VMware vCenter Operations Manager and services to manage everything from one control pane, and we can actually deliver our service in almost a one-click type of scenario. This allows us to on-board a new VAR or managed service provider to our WhiteCloud IaaS offering in a matter of hours.
NetApp: What led PeakColo to become an early adopter of NetApp unified scale-out storage and the Data ONTAP® 8.1 operating system?
PeakColo: We adopted the Cluster-Mode technology because it met our business needs. The ability to look at a single management pane and balance workloads across the infrastructure without any impact on our customers was a game changer for us. Prior to that, we had many small, discrete SANs that were a challenge to manage. It was difficult to scale our storage systems as fast as our business was growing and still deliver the 100% uptime SLA that is so critical to our customers.
Scale-out storage is the most powerful way to respond to data growth and data management challenges. Learn more today.
NetApp: How has the choice of storage platform affected your business? Have you been able to quantify an economic impact?
PeakColo: The use of Cluster-Mode has changed the economics of the storage services we can deliver to our customers, whether they use one terabyte or hundreds. We’re able to keep our head count down through automation. We enable our VARS to manage deduplication and compression for their own services and then pass those savings on to customers. With space savings as high as 60% to 70%, it can amount to the equivalent of one free terabyte for every two purchased. We also plan to extend the life of some of our older storage gear by adding it to the cluster and then using it to provide a lower class of service for noncritical storage, which is an economic benefit we hadn’t anticipated.
NetApp: What are some of the Data ONTAP technologies you’re considering for the future?
PeakColo: We’re currently testing some of the new capabilities supported in the Data ONTAP 8.1.1 release. We’re really excited about Flash Pools, which will allow us to use flash selectively at an aggregate level and maintain a very high level of performance when moving vServer containers between nodes in a cluster. We’re also looking at how we could implement Infinite Volumes to provide new service offerings for our big data customers that need a level of scale and performance that they can’t get with NFS today.
© 2013 NetApp, Inc. All rights reserved. No portions of this document may be reproduced without prior written consent of NetApp, Inc. Specifications are subject to change without notice. NetApp, the NetApp logo, Go further, faster, and Data ONTAP are trademarks or registered trademarks of NetApp, Inc. in the United States and/or other countries. Microsoft is a registered trademark of Microsoft Corporation. Oracle is a registered trademark of Oracle Corporation. SAP is a registered trademark of SAP AG. All other brands or products are trademarks or registered trademarks of their respective holders and should be treated as such.