Cloud and, to some degree, virtual form factors present exceptional
scalability as resources can be provisioned dynamically in line with
demand. Encoders are not tied to audience demand, contrary to
streaming servers, CDNs, etc that need to scale dynamically according
to service popularity. Encoders only need to scale based on the volume
of source content (live or file assets). Therefore, dynamic tasks such as
file-to-file transcoding or short-term events encoding typically run on
virtual or cloud form factors, while 24x7 linear encoding tasks do not
need such scalability.
PORTABILITY AND HIGH AVAILABILITY
Consumers have evolved their expectations for live and on-demand
media as a persistent, instant supplement to their subscribed services.
High availability is gaining importance for high-profile live events (e.g.
the World Cup), as well as for 24x7 linear TV services (e.g. AT&T U-verse
or BBC iPlayer). In both cases, stakes are high in terms of maintaining
operational efficiency without jeopardizing service uptime.
For video-on-demand transcoding, service reliability is less
important, as failed transcoding jobs are easily re-run. Thus,
relinquished control and relaxed high-availability constraints are
typical for such workflows, as long as video quality is not affected.
Operators often choose to utilize a virtual or cloud infrastructure for
such tasks, and file transcoding-as-a-service became popular early
on. The portability advantage of the cloud and virtual form factors
contributes further here, as operations can be moved closer to where
respective assets are stored or consumed.
For live encoding, however, reliability is key, as brand and customer
loyalty quickly dilute when sub-par performances of live streams are
delivered. Tier 1 telcos, content providers, and high-profile events are
thus less likely to relinquish control and responsibility to a third party
and generally favor turnkey or virtual form factors. As much as cloud
service providers claim high availability as an inherited feature of their
architecture, recent system-wide failures have scared Tier 1 operators
away from sole dependency on the cloud. Such operators turn to a
hybrid architecture based on a mix of turnkey or virtual encoding form
factors, with a disaster-recovery 1: 1 site on virtual or cloud ones. Less
strategic services, or Tier 2/3 operators with lower uptime constraints,
may try to avoid the additional CapEx element for their disaster
recovery site and rely on the cloud to dynamically provision services,
in an N+M fashion, in case of a failure at their primary site.
Nevertheless, high availability is delivered end-to-end and not in
isolation; thus, the optimal solution is dictated by the entire ecosystem
and not just the encoding component. However, this does not mean all
components of the ecosystem need to have the same form factor.
Finally, often overlooked is the human factor. The history, culture
and practices of the organization play a substantial role in the success of
a project, especially as the three form factors expect diverse skill sets.
Turnkey appliance form factors fit well with engineering teams
that are video-oriented and/or organizations that focus on core
competences while maintaining ownership of their operations. The
virtual form factor demands strong IT-oriented engineering skill sets
and fits organizations that value flexibility and tighter control. Finally,
the cloud form factor typically suits teams/organizations comfortable
with delegating control (and responsibility) to third parties, often for
short-term or less strategic aspects of their operations.
Clearly, none of the three form factors is a panacea for multiscreen
encoding. The operating, strategic and competitive practices of each
organization dictate the appropriate architecture. Furthermore, it is
often fitting to consider a solution that spawns across multiple encoding
form factors, either simultaneously or as operations evolve. In order
to deploy such hybrid and fluid architectures, choosing an encoding
solution that is present, proven and seamless across all form factors is
essential. Media Excel’s HERO product portfolio presents an established
solution along those objectives and is proven in deployments by Tier 1
telcos, broadcasters and content providers worldwide.
ABOUT MEDIA EXCEL
Founded in 2000 and headquartered in Austin, Texas, the company
has been the industry leader in the development of real-time video
processing software to distribute video over IP networks. Media
Excel’s products provide the reliability, scalability and performance
required to deliver high-quality video via turnkey, virtual and cloud
architectures. Further, Media Excel’s orchestration and telemetry
ecosystem enables operators to deploy services across multiple
form factors and seamlessly migrate services across them based
on pricing, strategic or tactical stimuli. Serving more than 500
million multiscreen subscribers around the globe, and with the #1
market share in multiscreen delivery for wireless carriers in North
America, Media Excel powers Pay TV operators, content providers,
broadcasters and telcos worldwide. To learn more, please visit
www.mediaexcel.com or contact us at firstname.lastname@example.org.