Winter/Spring 2026 - Journal - Page 27
IT & IT SECURITY
the problem. We work with
customers whose networks
boast Gigabit symmetrical
internet but struggle with poor
response times for users. We
see clients with low throughput
connections who consistently
achieve low latency rates.
Understanding the difference
and knowing which levers to pull
can save you countless hours
of troubleshooting and actually
improve application performance
where it matters.”
Exploring the options
So, what can storage teams do
when they wish to tackle and
mitigate the effects of latency and
packet loss? WAN Optimisation
is one option, but it can’t send
and receive encrypted data.
Then there are SD-WANs, but
they often need a boost, and
that often comes by creating a
WAN Acceleration overlay. WAN
Acceleration is a technology,
which isn’t to be confused with
WAN Optimisation, which
deploys arti昀椀cial intelligence
(AI), machine learning and data
parallelisation to increase data
transfer rates and improve
bandwidth utilisation by up to
98% without needing to buy new
gear.
As I mentioned to trade show
blog, Cloudfest, a couple of
months ago, ‘The Cloud Latency
and Packet Loss Guide: How to
Fix Them?’ – written by freelance
business and technology
journalist, Graham Jarvis, there
is also a need to consider
the impact of centralising the
corporate business and service
continuity capabilities. By putting
them in the same circles of
disruption to reduce latency,
these very important facilities
when they are brought down by,
for example, a cyber-attack or a
power outage within the same
local vicinity are put in jeopardy.
Potentially they could impact
uptime.
Latency and packet loss
damage
“Packet loss is especially
damaging in encrypted
communications or 昀椀nancial
transactions, where missing data
may trigger retransmissions or
session failures, undermining
both performance and trust,”
writes Niksun in its blog, ‘Latency,
Jitter, and Packet Loss: The
Key Metrics De昀椀ning Network
Performance Management.’
“Hosting within the same data
center, or shadow center, is still
an issue. Sometimes failover
zones are created by splitting the
data center in half, but it’s only
昀椀ne if you don’t lose power to
both sides. If you really want to
have a secure disaster recovery
system, you need a 3-2-1-1-0
approach.”
Jarvis writes: “WAN
Acceleration can send and
receive voluminous amounts of
encrypted data. It is also data
agnostic, and it can’t be seen
by those, such as Bridgeworks,
providing the service.” Yet the
challenge can often be about
regulatory compliance – pushing
disaster recovery sites and
data centres closer together
to ensure compliance to other
considerations.
Reducing total storage and
network costs
Thankfully, WAN Acceleration
can reduce total storage and
network costs, while improving
Recovery Point Objectives (RPO)
and Recovery Time Objectives
(RTO). A Formula 1 company, a
household named cyber-security
company, and the likes of Veeam
have experienced its bene昀椀ts in
the improvements it can offer.
This includes faster backups
and restores – making business
and service continuity more likely
in the event of a natural, cyber
or human-created disaster. So,
rather than treating the WAN as
a 昀椀xed limitation, this approach
can unlock better returns from
existing storage and backup
investments and make poor
WAN performance in backup and
replication a thing of the past.
David Trossell
Problems with regulation
In the article, he points out that
the EU’s Cloud Sovereignty
Framework may require data
to be stored within national or
regional boundaries. Despite
these requirements, distance
has its advantages – particularly
when data centres and disaster
recovery sites are located at a
distance and outside of each of
their circles of disruption. The
issue is that latency becomes
more problematic the farther
away you transfer data. At
present, the faster was to achieve
this is with WAN Acceleration.
WAN Acceleration’s role
is therefore to expedite data
transfers and to safeguard it,
while improving offsite backup
and remote replication ef昀椀ciency.
The alternative is not worth
being complacent about. Niksun
explains in its blog: “Latency,
jitter, and packet loss don’t
just affect user experience;
they directly impact business
continuity. High packet loss
during a live customer support
session or excessive jitter during
a CEO webcast can damage
brand credibility. Moreover,
these issues often signal deeper
vulnerabilities — such as
miscon昀椀gurations or potential
security threats.”
GOVERNMENT AND PUBLIC SECTOR JOURNAL SPRING/SUMMER 2026
27
