In Broadcast - October 2015 - 54

www.inbroadcast.com | Vol: 5 - Issue 10 | October 2015

InMedia

JITT Favoured Over JITP
For Live-To-VOD At The Edge
In Media Processing Wars
By Steve Sklepowich, Vice President
of Marketing at Vantrix

perators are unsure how to cope
with the explosive growth in timeshifted and place-shifted consumption
of live, linear content. Projecting 250%
growth in 2015, research firm Frost
& Sullivan recently put the two top
transcoding workflows to the test. What
they found out after comparing the
TCO of the "incumbent" just-in-timepackaging (JITP) against "newcomer"
just-in-time transcoding (JITT), was
that live-to-VOD media processing
capability at the network edge will
better enable operators to compete
and thrive.
In this article, we'll summarise
some of the research group's most
important findings. The full report can
also be downloaded at this page.
Frost & Sullivan considered two
factors when modelling the CAPEX for
live-to-VOD at the edge: cost of storage
and cost of transcoding. For simplicity,
analysts made several assumptions,
such as that storage is deployed on
premises.
The cost of storage scales
approximately linearly with the
quantity of content that must be
stored. For Network DVR or nDVR
applications, current copyright law
interpretation in many jurisdictions
- including in the United States requires that one copy of content be
stored per user, rather than allowing
all playback requests to be served
from the same copy. In the JITT case
an operator will require approximately
20TB of storage per 1,000 subscriber
households.
In contrast, JITP requires a capacity
of approximately 60.5TB on average

Relative TCO and Ongoing Savings

for the same parameters. The CAPEX
for storage in the JITP case is slightly
less than 3X the CAPEX for the JITT
alternative, since the per-unit cost
of storage decreases with increased
volume.
The specifications of the transcoding
horsepower necessary for the JITT
versus JITP case vary tremendously,
and thus the cost of the JITT transcoder
is 10X the cost of the JITP alternative.
However, this higher expense is more
than offset by the CAPEX savings on
the storage side.
OPEX considerations consist of
the run-time cost of the equipment,
including power, rack space overhead
and personnel oversight requirements.
They assume an operating expense
rate of $0.03/MB of transcoding for
the higher-intensity JITT case, and
half that rate for the less demanding
JITP case. All recorded content is
transcoded into a number of preset
profiles in the JITP scenario. However,
for JITT, only content actually
requested is transcoded, and only at
one given profile. Statistics show that
subscribers only view about a fourth
of the content they record. This has
significant, positive implications on the
transcoding workload in the JITT case.
The charts below summarise the
differences in the design parameters
and consequently the differences
in upfront investment and ongoing
expenses between JITP and JITT.
Frost & Sullivan found that CAPEX
savings are relatively modest at
about 30%, while ongoing OPEX
savings are compellingly significant at
approximately 60% every year. The

Relative Storage, Relative Transcoding Volume and Relative CAPEXOPEX

key differences in CAPEX come from
the variation in storage requirements
and transcoding volumes in the JITT
and JITP use cases in the base year,
current cost of storage, and current
costs of appropriate transcoders and
packagers.
Under a conservative forecast model,
assuming that ongoing expenses will
rise moderately as volumes rise but
technology costs will fall, Frost &
Sullivan found that the TCO for JITP
is approximately twice that of JITT.
The relatively high price point of a
JITT transcoder relative to a softwaredefined transcoder developed for JITP
purposes can give operators pause at
first look. However, the analysts found
that when the cost of a JITT transcoder
is weighed against the net present
value (NPV) of OPEX savings generated
over a five-year period, the return on
investment is over 200%.
The analyst's key takeaway is that
operators can significantly reduce
upfront and ongoing costs for live-toVOD at the edge by choosing a state-ofthe-art JITT architecture, as compared
to a JITP deployment. Another way of
interpreting these findings is that for
the same investment levels, JITT allows
operators to build out significantly
higher capacity and deliver much
better quality of experience to
subscribers.
The analysts examined the Vantrix
Media Platform (VMP) as an example

of an ultra-high-density transcoder
enabling JITT scenarios for operators.
VMP is a software-defined solution
that enables video service providers
to cost-effectively deliver high-quality,
multiscreen video. VMP can be
deployed on ultra-high-density turnkey
video processing appliances, or on
standard servers in private or public
cloud scenarios.
With this approach operators
are able to combine efficient IP
conversion,
ultra-high-density
transcoding, adaptive bitrate (ABR)
packaging, encryption and streaming
into a modular, virtualised solution.
The result is a highly flexible and costefficient way to give consumers highquality video today with a future-proof
feature set. Customers can achieve up
to 80% cost reduction per stream by
deploying Vantrix Video Processing
modules on industry-leading 4.3U,
2U and 1U appliances. These turnkey
solutions also reduce footprint by 95%,
resulting in dramatic CAPEX and OPEX
savings.
From a long-term opportunity and
competitive differentiation perspective
Frost & Sullivan delivers powerful
arguments for revisiting the current
norm of relying on just-in-time
packaging. Download the full report to
get an even deeper look at the reasons
why the economic model comes out
emphatically in favour of JITT at the
network edge for live-to-VOD.

http://www.inbroadcast.com http://vantrix.com/home/ http://vantrix.com/home/

Table of Contents for the Digital Edition of In Broadcast - October 2015