VMTurbo recently launched version 3.0 of its award-winning platform, VMTurbo Operations Manager
3.0. To find out more, I spoke with Yuri Rabover, the company's co-founder and VP of product strategy. Here is that conversation:
VMblog.com: Can you tell us a little bit about VMTurbo and how the company was founded?
Yuri Rabover: VMTurbo
delivers an intelligent workload management solution for cloud and
virtualized environments with the goal of helping customers get the most
out of - and continue to expand - their virtualization deployment. The
founding team of VMTurbo came from SMARTS, an innovator in system and
network root cause analysis solutions (acquired by EMC in 2004). The
core founding team started VMTurbo in 2009, realizing that the rapid
growth of virtualization in the datacenter was redefining the management
requirements for companies of all sizes. VMTurbo is backed by Bain
Capital Ventures and Highland Capital Partners. Today, the company is headquartered in Massachusetts, with offices in New York, California, United Kingdom and Israel.
VMblog.com: What does your company mean when it talks about "intelligent workload management"?
Rabover: We
believe that intelligent workload management provides an opportunity to
profoundly change the way datacenters are managed - one that is based
on system intelligence as a whole and leverages the fluidity that
virtualization provides to continuously tune the environment and assure
application performance while also utilizing the underlying resources as
efficiently as possible. It's not about threshold management,
predictive analytics, alert suppression and root cause diagnostics. It's
about intelligently preventing problems from occurring in the first
place - by optimizing and removing contention across the entire
environment.
VMblog.com: So what's the secret to doing this right? How does VMTurbo pull it off?
Rabover: The
essential question IT is trying to solve for in a virtual environment
is: how do I assure and prioritize application performance while also
getting the absolute most out of my shared physical resources? Answering
this is obviously not a static proposition; it's constantly evolving as
demand fluctuates. Our solution takes a fundamentally different
approach to solving for it - one that does not focus on threshold
management for the thousands of control points in a complex virtualized
datacenter. Instead, we represent the IT environment as a market - where
the physical resources (storage, compute, memory, network I/O) are
represented as "supply" and the virtual workloads and applications that
run within them are represented as "demand". When you look at the IT
environment this way, and allow for the market to "self-level" based on
availability of budget and supply, you make better decisions about what
workload to run where, and how to continuously tune the environment to
avoid resource contention and performance problems. This mirrors how
buyers continuously seek better prices and suppliers strive to maximize
profit in a free market. These decisions are entirely automatable
because you are addressing potential issues before the environment is
stressed. It also provides a better way to perform what-if scenarios as
operators, architects and application owners assess virtualizing
critical applications, adding new users or upgrading server hardware.
VMblog.com: Your company recently announced version 3.0 of its platform, what are the new features in this release that people need to be aware of?
Rabover: The new capabilities in VMTurbo Operations Manager 3.0 include:
1. Cloud-scale management through
VMware vCloud Director integration. vCloud Director (vCD) enables the
consolidation of virtual infrastructure across multiple clusters or
virtual datacenters in a multi-tenant environment. Our product is able
to manage resource allocations within these datacenters and across the
vCD topology, ensuring application performance and avoiding the
oversubscription and resource contention that often occurs with the
added layer of abstraction that vCloud Director creates in large-scale
environments.
2. Single-instance, multi-hypervisor management
by adding Citrix XenServer support and providing IT operators with the
unique ability to manage performance and control configuration across
XenServer, vSphere and Microsoft Hyper-V environments from a single
console.
3. Application-aware service assurance
that now discovers and profiles Windows applications across the virtual
environment providing visibility into application performance and
enabling IT operators to prioritize resource requirements for critical
applications - ensuring quality of service as demand fluctuates.
4. Enhanced, intelligent capacity planning
which utilizes the product's deep understanding of workload performance
characteristics and the physical environment to conduct
performance-based capacity analysis.
VMblog.com: What are some of the biggest management challenges that enterprises are currently facing in their data centers?
Rabover: Across
the board our customers tell us that too many cycles are spent finding
and trying to fix discrete performance problems, instead of preventing
them in the first place. Determining the proper reaction to problems in a
virtualized environment is very complex after the fact. Threshold-based
management only exacerbates the problem - even if the thresholds are
"intelligent" or "predictive" as many solutions purport. It requires a
high degree of skill and domain knowledge and is very difficult to
automate. Enlightened IT operators understand that virtualization
provides the fluidity required to prevent contention and performance
issues, but it also requires a different approach to management. When
done correctly, virtualization is automatable and places control back
into the hands of customers by intelligently tuning the configuration of
the virtual environment to provide application assurance and avoid
resource contention.
###
Once again, thank you to Yuri Rabover, co-founder and VP of product strategy at VMTurbo, for taking time out to speak with VMblog.com.