Skip to main content
← All videos

Linear Scaling Benchmark — 1 → 2 → 3 Nodes, Live

Watch throughput climb in real time as nodes are added — 630K to 1.3M ops/sec, one workload, on the real console.

5 minWatch on YouTube ↗

Watch it, or read it

The full walkthrough below — every screen from the video, with the narration transcribed. Skim it, search it, or read it in your own language.

Linear Scaling Benchmark — 1 → 2 → 3 Nodes, Live — screen 1
A live scaling benchmark of Zaris, on the real management console. We start with a single store node, switch on a fleet of client applications, and watch throughput climb as we add nodes — one, then two, then three. Clients generate the traffic; store nodes serve it.
Linear Scaling Benchmark — 1 → 2 → 3 Nodes, Live — screen 2
One node, under full load. The client applications are firing reads at it, and the throughput tile settles around six hundred and thirty thousand operations per second — a single four-core machine running at nearly one hundred percent CPU. This is the ceiling for one node.
Linear Scaling Benchmark — 1 → 2 → 3 Nodes, Live — screen 3
To go faster, we do not touch that node — we add another. A second store node joins the cluster, and Zaris automatically rebalances, moving partitions onto it so the work spreads evenly across both machines. No manual resharding, no downtime.
Linear Scaling Benchmark — 1 → 2 → 3 Nodes, Live — screen 4
With two nodes sharing the same workload, throughput climbs to about one million operations per second. Nodes online: two of two, both running, both pulling their weight — and all we did was add one machine.
Linear Scaling Benchmark — 1 → 2 → 3 Nodes, Live — screen 5
Now a third node. Again it joins, partitions redistribute across all three, and — because it takes more traffic to keep three nodes this busy — we bring in a few more client applications to generate the load.
Linear Scaling Benchmark — 1 → 2 → 3 Nodes, Live — screen 6
Three of three running, and there is the payoff: throughput climbs past one point three million operations per second, peaking near one point four. From one node to three, the cluster scaled almost linearly, every node carrying its share.
Linear Scaling Benchmark — 1 → 2 → 3 Nodes, Live — screen 7
Add nodes, get throughput — with automatic rebalancing, no downtime, and no manual resharding. That is horizontal scaling in Clustron Zaris. Head to clustron.io to explore the numbers and try it yourself.

More videos

Getting Started — a Distributed Cache for .NET in MinutesSecuring a Distributed .NET Cache — Authentication & RBACTLS & Mutual TLS for a Distributed .NET CacheDistributed In-Memory Cache Architecture, Explained
Get startedSubscribe on YouTube