Citi Prime Finance’s 2011 IT Trends & Benchmarks Survey
I
23
As mentioned earlier, funds of this vintage also took advantage
of the increased capacity and bandwidth to augment their
disaster recovery / continuity of business environments. Data
could now be replicated in shorter and shorter intervals from
production environments, and fail-over to these environments
became much more seamless, leading to minimal loss of data
and less downtime. Some of the earlier off-premises data
centers of this vintage weremaintained by the funds themselves;
having rented rack space directly from existing data centers, the
internal IT staff at the fund would travel to the remote location
to maintain all the hardware, from the operating systems on up
through the business applications.
Despite this overhead, managers were willing to undergo this
effort to mitigate the business risk of losing data and having
their networks – and therefore their trading operations – be
down for extended periods of time. Such controls began to
assuage investor concerns, as questions on disaster recovery
increasingly appeared within due diligence questionnaires.
Hedge Fund 3.0 Allows Managers to Move Their
Entire Infrastructure Off-Premises
Given the interest in adopting off-premises solutions by hedge
funds up to this point—by way of service providers such as
prime brokers and administrators, vendor-hosted software
and disaster recovery environments—it didn’t take long for
infrastructure frms to focus almost exclusively on the hedge
fund space to adopt the “managed service provider” model. As
noted earlier, in this model, IT infrastructure frms would rent
cages within established data centers and, in turn, these frms
would lease server capacity to hedge fund clients.
It is worth noting that these cloud environments are typically
hosted on discreet sets of servers—a “virtual private cloud.” This
approach differs from the generic cloud-based hosting services
model offered by frms such as Amazon and Google. In the
generic cloud-based model, data and services are distributed
across multiple, anonymous servers and CPUs, but in the Hedge
Fund 3.0 model, the managed service providers are typically
allocating specifc servers for each client.
Early pioneers of this model include Options IT in the UK and
InfoHedge in the U.S. More recently, other IT infrastructure
frms, including Abacus and Auxia, have launched similar
platforms, while the traditional providers from the frst and
second iterations of hedge fund infrastructure development
have adjusted their business models to refect this new reality,
offering traditional on-premises network build-outs and support,
as well as hosted solutions.
How Hedge Fund 3.0 frms deploy their software also evolves.
With the hosted infrastructure model, hedge funds now have a
new option on how to access software. They can opt to have
their vendor install their application within the hedge fund’s
own virtual private cloud at their managed service provider’s
data center. For some frms, this may be seen as offering more
security than choosing to access software-as-a-service directly
from the vendor’s own hosted platform.
In the vendor-hosted software-as-a-service model, applications
may be “multi-tenant,” with a fund’s data segregated by
permissioning within the application, not by way of segregated
servers. Delving into this point with various providers will be
necessary for those funds that take the most-conservative
approach toward data security although most software
providers who manage their own hosted environments are often
able to present strong enough security credentials to get many
funds comfortable.
“The IT department needs to provide the CFO with insight
into their costs and how those costs are beneftting the
organization. Since IT has a full-time job managing the
infrastructure, technology expense management is ripe for
business-process outsourcing, which can result in driving
costs down through exchange rebates and vendor contract
management.”
– CEO of Expense Management Business
Process Outsourcing Firm
“There are three ways to handle infrastructure: 1) hire
your own team and buy your own gear; 2) buy the gear and
rent the labor by outsourcing it to an integrator; or
3) outsource the hardware and the labor by leveraging a
managed service provider.”
– CEO of a Managed Service Provider
in the Alternative Asset Space