Infrastructure

Diagnose lag before it costs you players

When players say the server feels bad, infrastructure telemetry shows whether it is the tick loop, memory pressure, garbage collection, or the data pipeline. It is the operational depth generic analytics tools do not have.

What Infrastructure does for you

TPS, MSPT, JVM runtime, garbage collection, pipeline health, and network interfaces in one operational view.

TPS and MSPT

Track ticks per second and milliseconds per tick with the shape of the slowdown, not just the current value.

JVM runtime

Runtime, uptime, threads, and memory pools for the Java process behind your server.

Garbage collection

GC events and pause behaviour, so memory pressure is visible before it becomes a stall.

Pipeline health

See whether collection, ingestion, and processing are healthy, delayed, or failing.

Network interfaces

Traffic and interface metrics from the host, alongside the player-facing latency view.

Worlds and spatial

Player distribution and activity across worlds and regions for server-scoped analysis.

How you'd use it

Three real situations owners and staff run into, and what infrastructure answers.

TPS drops after an update

The situation

Players report lag immediately after a plugin release.

With ShardAnalytics

Line up the TPS drop with the deployment window, check JVM memory and GC, and identify the pressure pattern before rolling back blindly.

Memory pressure at peak

The situation

The server is fine all day and struggles only at peak hours.

With ShardAnalytics

Memory pools and GC pauses show whether the heap is the constraint and whether a restart pattern is hiding it.

Collection looks wrong

The situation

The dashboard numbers seem stale but nothing is obviously broken.

With ShardAnalytics

Pipeline health and readiness distinguish a stalled collector from a slow query, with the missing stream named.

How it works

Install the connector

The plugin reports performance telemetry alongside player and event data.

Open a server's observability

Read TPS, MSPT, JVM, GC, and pipeline health for that server.

Alert on the bounds

Turn the thresholds you care about into alerts so you hear about them first.

Questions

Does this replace a profiler?

No. It gives you the operational timeline and context to know when to profile, and it keeps that history next to player impact.

Can I alert on these metrics?

Yes. TPS, MSPT, latency, CPU, and memory thresholds can all drive alerts to Discord or email.

Start free during the beta

Launch your first dashboard, monitor your servers, and explore real-time analytics. Every feature is free while ShardAnalytics is in open beta — no credit card required.