Skip to main content
PKB Solutions logo

Working with us

Start with a scoped conversation

Tell us the outcome you are chasing and we will map it to a realistic first phase.

Contact the team
PKB Solutions logo
Home

Infinity Platform

VDI Experience

A virtual desktop can be perfectly healthy at the broker and unusable at the desk. VDI Experience measures from inside the session, so you find out which one you have.

PKB VDI Experience console showing 3,480 active sessions, round-trip latency, logon duration and host capacity headroom per pool.

What it covers

Four things a broker cannot tell you

Broker health confirms a session was handed out. It says nothing about what happened next.

Protocol responsiveness

Round-trip time, frame delivery and input lag measured inside the session rather than at the connection broker.

Logon duration, decomposed

Broker time, authentication, host allocation, profile load and shell readiness separated so the slow phase is identifiable.

Host and pool capacity

CPU, memory, storage and GPU headroom under real concurrency, not against a provisioning spreadsheet.

Location and last mile

Home, office and client-site conditions separated, so a site problem is not mistaken for a platform problem.

Capabilities

Discover key features

Three surfaces your virtual desktop team will actually live in.

Session experience table listing the six worst locations by measured round-trip time, packet loss and health state.
01

Protocol quality

See the session the way the employee sees it

Latency, loss and responsiveness measured per session and grouped by location, so you can separate the network from the host without guessing.

  • Round-trip and packet loss per location, not per datacentre
  • Home users separated from office and client sites
  • Worst-affected locations surfaced first, not buried in an average
Capacity view showing concurrent sessions by hour of day against provisioned capacity, and host utilisation meters at peak.
02

Capacity

Size the estate against measured demand

Concurrency by hour against what is actually provisioned, with host utilisation at the peak window. It is the difference between buying capacity and buying confidence.

  • Concurrency curve against provisioned capacity
  • CPU, memory, IOPS and GPU headroom at peak
  • Pools ranked by how close they run to the edge
Logon flow diagram from client through broker and authentication to host allocation, profile load and shell ready, with a timing tile for each stage.
03

Logon

Find where the thirty-one seconds go

Logon is decomposed into its stages and each is timed separately, so the conversation moves from “VDI is slow” to “profile load is fourteen seconds and here is why”.

  • Broker, authentication, allocation, profile and shell timed apart
  • Directory latency isolated from host allocation
  • Change events overlaid on the timeline

Common confusion

Three problems that look identical from the broker

All three present as “the session is slow”. Each needs a different team to fix it.

Three problems that look identical from the broker
Symptom as reportedActual causeWhere the evidence sitsWho owns the fix
Session feels laggyLast-mile latencyIn-session round-trip by locationNetwork
Session feels laggyHost CPU saturationUtilisation at the peak windowPlatform
Session feels laggyApplication inside the sessionPer-application responsivenessApplication owner
Logon takes foreverProfile size growthProfile load phase timingWorkplace
Logon takes foreverDirectory response timeAuthentication phase timingIdentity

Bring us a pool that is under-performing

Pick the pool your users complain about most. We will walk through exactly which phase we would measure first.