Slide 27
Slide 27 text
Single Server Hyperscale (Citus)
Basic General Purpose Memory Optimized Coordinator node Worker node
Intended
workloads
Light compute and
variable I/O performance
Balanced compute and memory
with scalable I/O throughput
In-memory performance
for faster processing
and high concurrency
Highly scalable performance for large data sets,
complex queries, and high concurrency
vCore 1, 2 2, 4, 8, 16, 32, 64 2, 4, 8, 16, 32 4, 8, 16, 32, 64 4, 8, 16, 32, 64
Memory 2 GB per vCore 5 GB per vCore 10 GB per vCore 4 GB per vCore 8 GB per vCore
Storage 5 GB – 1 TB
Magnetic Storage
5 GB – 16 TB
Remote SSD
5 GB – 16 TB
Remote SSD
0.5 – 2 TB
Remote SSD
0.5 – 2 TB
Remote SSD
IOPS Variable 3 IOPS/GB 3 IOPS/GB 3 IOPS/GB 3 IOPS/GB
Hyperscale (Citus) deployments include one
coordinator node and at least two worker nodes.
Scale horizontally by adding worker nodes.