Running
—
Workloads
Start Latency
—ms
Avg cold start
CPU
—%
128 cores @ peak
Memory
—GB
of 512 GB
Token/s
—M
LLM throughput
Launches Today
—
cold starts
Errors
—
Need attention
Workload Distribution
Status Matrix · —
Type
Count
Refresh
Workload Events
Workload Events
LIVE
⚡ COLD-START SPEED
—ms
Avg cold-start latency · Goal <100 ms
0ms—Traditional 2000 ms
Core Performance Metrics
EMON / Perf
CPI
—
cycles / instr
L3 Miss
—
LLC miss rate
Mem BW
—
DRAM bandwidth
CPU—
Memory—
Network—
Workload Details