Most organizations today have embraced virtualization technology because it significantly reduces the number of physical servers they need to maintain while providing the increased flexibility of an abstracted environment. While this consolidation of physical servers helps reduce data center costs, it also introduces new challenges and complexity in managing a virtualized environment and its impact on networked storage.
Assessing VM Needs
With ten, twenty or possibly even hundreds of Virtual Machines (VM) running within one physical server, IT now has to be concerned with ensuring that each of those VMs has the necessary resources to avoid impacting the performance of the applications running on each of those VMs. In order to accomplish that, admins need to be able to determine accurately things like compute, memory, performance and storage resources, as well as workloads on each VM. While there are a variety of utilities and tools to discover this information, having to do it manually on all VMs in the environment is inefficient and prohibitively time consuming. Additionally, most of these tools and utilities do not provide the admin with a clear view of what is actually happening in their storage systems, nor the impact on individual VM workloads.
Visualizing VM Needs
What is needed to address these challenges is a solution that is able to make the enterprise “data aware” by gathering file level metadata from the clients to provide detailed insight into data, including the ability to monitor performance, workloads and availability levels of the VMs and applications as well as the status of all storage systems. Beyond data awareness, a number of other elements ensure that such a solution will streamline how organizations can automatically meet VM needs.
In modern IT, time is of the essence, and nothing communicates faster than a graphical representation of the current condition. As such, an ideal solution for managing VM needs should be able to display all of this information for each VM and the storage environment, in a graphical format via a GUI that is easy to customize and makes the information easy to understand.
The solution should also be policy driven in order to enable the organization to create data policies that ensure their Service Level Objectives (SLOs) across performance, latency, cost and protection are properly matched with the underlying storage capabilities in order to optimize storage efficiency and the application user experience. By automatically aligning data to the right type of storage for VM needs, organizations can avoid the need for a manual Storage vMotion in VMware vSphere. This again minimizes the time spent on ensuring IT resources are aligned with changing business needs.
Additionally, the solution should fully support VMware VVOLs by acting as a universal VASA provider, which enables existing storage resources, from server-side flash to NAS and cloud storage in the enterprise, to be presented as a single, easily managed VVOL datastore, irrespective of storage vendor brand, management, or protocol type. This lets organizations adopt VVOLs without the need to purchase new VVOL-certified storage resources.
An ideal solution would also be able to automatically analyze VM performance and workloads, then automatically balance workloads across storage resources by dynamically moving VMDKs across heterogeneous storage in response to changing SLOs, and VM requirements to ensure VMs meet IT-defined SLOs. It would also be able to automatically move VMs to the ideal server for compute or availability needs.
Another necessary feature would be the ability to create point-in-time copies of the VMDK that can be moved like backups to more cost effective storage media than the primary copy, including storage systems with deduplication capabilities, as well as object storage and public cloud storage using standard S3 or Swift interfaces.
StorageSwiss Take
Faced with limited staff, flat budgets, ever-increasing amounts of data, and rapidly growing virtual environments, organizations today need an effective, vendor agnostic, data management solution that provides them with a clear insight into all aspects of their storage and virtual environments. This solution also needs to provide enterprises with the means to automate their storage management functions based on data policies and SLOs defined by the organization. This will make the most cost effective use of existing resources while helping to reign in storage sprawl.
Joseph is a Lead Analyst with DSMCS, Inc. and an IT veteran with over 35 years of experience in the high tech industries. He has held senior technical positions with several major OEMs, VARs, and System Integrators, providing them with technical pre and post- sales support for a wide variety of data protection solutions. He also provided numerous technical analyst articles for Storage Switzerland as well as acting as their chief editor for all technical content up to the time Storage Switzerland closed upon their acquisition by StorONE. In the past, he also designed, implemented and supported backup, recovery and encryption solutions in addition to providing Disaster Recovery planning, testing and data loss risk assessments in distributed computing environments on UNIX and Windows platforms for various OEM's, VARs and System Integrators.
StorageSwiss.com - The Home of Storage Switzerland 