Virtualization and Cloud executives share their predictions for 2016. Read them in this VMblog.com series exclusive.
Contributed by Jim Frey, the VP of Product Management at Kentik
ITOM Will Grow Up and Integrate
One
of the sectors of the IT industry that has been least touched by cloud
computing is IT Operations Management (ITOM) tools. Beyond Application
Performance Management (APM) and
wireless LAN management, most ITOM practices are highly siloed,
engineer-centric, and based on-premises enterprise software and
appliance tools. This is starting to change, and we'll see that change
continue and accelerate in 2016.
Cloud
computing has made an indelible mark on IT practices. That IT will
operate hybrid clouds with a heavy and perhaps dominant dose of public
IaaS and SaaS is unquestioned at
this point. Gartner reports that most net new applications are being
developed for public cloud deployment. API-friendliness is de rigueur
in this new reality. Accordingly, provisioning and configuration
management has moved towards API automation model.
Chef, Puppet, Ansible and Saltstack are great example of this movement.
Accompanying
this cloud trend is a DevOps movement with roots in agile development
and Lean IT. DevOps largely presumes both an API-driven infrastructure
and high degrees of automation.
It adds to that an organizational culture value for cross-team
collaboration. From development through deployment, App Dev and IT Ops
teams are expected to work in tandem to practice continuous integration,
continuous delivery and continuous deployment.
Compared
to the progressiveness of the cloud and DevOps trends, ITOM is a bit of
a laggard. Most ITOM practices are still built on organizational
hierarchies and "throw it over
the wall" ways of doing problem escalation. Tools are still primarily
deployed as point solutions or as monolithic enterprise software. Most
of today's dominant legacy ITOM tools were architected in an older,
scale-up model rather than a cloud, scale-out
model. As a result, in many cases, the high volume of analytics and
diagnostics data that IT infrastructure produces can't be stored, so
much of it is thrown out and only summary snapshots of the data are
retained. And while some tools can handle the load,
deploying enough hard infrastructure to get there becomes prohibitively
costly to deploy and maintain. Tool interfaces and practices are still
primarily GUI-based, and it can be argued that a lot of ITOM
institutional knowledge consists of "muscle memory"-knowing
where to click a mouse on a screen. APIs in the ITOM space have not
been used for direct end-user interaction so much as vendor and custom
software integration, commonly at high services costs and commonly
resulting in a brittle architecture that is difficult
to move forward when new releases (and new feature capabilities)
arrive. As a result, valuable analytics data is often trapped inside
these tools, severely limiting or effectively halting cross-team
collaboration, automation and efforts to achieve continuous
process improvement. The focus is on survival.
This
model is clearly unsustainable in a cloud and DevOps world. We've
already seen a movement towards new, open-source tools like the ELK
stack. We've also seen more open, API-lead
ITOM SaaS startups, and 2016 should bring more. These are healthy
developments and we're rapidly approaching a tipping point where it will
no longer be justifiable to continue to invest in legacy tools and
build siloed practices around their rigid GUIs.
The major ITOM vendors have seen the writing on the wall and are
working to shift their solutions to a hybrid cloud model. APM will
continue its already significant shift into the cloud. Systems
management cloudification has started, and will ramp. Catalyzed
by interest in software-defined architecture, network management, often
the last to join every phase of IT modernization, will cloudify rapidly
starting in 2016.
Cloud
is a disruptive force, because it upends a key piece of IT's historical
identity-which is owning, configuring and maintaining infrastructure
and applications. Cloud is also
inevitable. ITOM is a latecomer to the cloud game, but as it starts
down that road, we'll see similar kinds of disruption in practices,
tools and vendor ecosystem as we've seen in other sectors. 2016
promises to be an interesting year for ITOM!
##
About the Author
Jim Frey, VP Product Management
Jim
has worked at the forefront of IT and network management technology for
over twenty years. His experience includes work as an industry analyst
at Enterprise Management
Associates (EMA), executive roles at NetScout and Micromuse, and
product marketing roles at Agilent and Cabletron. As VP of Research for
EMA, Jim covered the changing demand landscape for enterprise network
and infrastructure management tools, technologies,
and practices via direct primary research, practitioner dialogue, and
technical product studies. As VP of Marketing at NetScout, Jim also
played a key role in planning and executing joint marketing projects
with NetScout's major alliance partners, including
HP, Mercury, Riverbed, and OPNET. He also acted as communications
director for NetScout's acquisition and integration of Network General
and the Sniffer line of performance management and troubleshooting
products. Throughout his years in the infrastructure
and network management sector, Jim has been a regular speaker at
industry events around the globe and has been published in several
journals, proceedings, magazines, and textbooks.