Nova flavors
In NERC OpenStack, flavors define the compute, memory, and storage capacity of nova computing instances. In other words, a flavor is an available hardware configuration for a server.
Currently, our setup supports and offers the following flavors
NERC offers the following flavors based on our Infrastructure-as-a-Service (IaaS) - OpenStack offerings (Tiers of Service).
1. Standard Compute Tier
The standard compute flavor "cpu-a" is provided from Lenovo SD530 (2x Intel 8268
2.9 GHz, 48 core, 384 GB memory) server. The base service unit (SU) is 1 vCPU, 2
GB memory at a rate of $0.016 / hr of wall time. Multiples of the cpu-a SU are
available with 20 GB root disk, and the price scales accordingly:
| Flavor | SUs | Cost / hr |
|---|---|---|
| cpu-a.1 | 1 | $0.016 |
| cpu-a.2 | 2 | $0.032 |
| cpu-a.4 | 4 | $0.064 |
| cpu-a.8 | 8 | $0.128 |
| cpu-a.16 | 16 | $0.256 |
2. Memory Optimized Tier
The memory optimized flavor "mem-a" is provided from the same servers at "cpu-a"
but with 8 GB of memory per core. The base service unit (SU) is 1 vCPU, 8 GB
memory at a rate of $0.030 / hr of wall time. Multiples of the mem-a SU are
available with 20 GB root disk, and the price scales accordingly:
| Flavor | SUs | Cost / hr |
|---|---|---|
| mem-a.1 | 1 | $0.030 |
| mem-a.2 | 2 | $0.060 |
| mem-a.4 | 4 | $0.120 |
| mem-a.8 | 8 | $0.240 |
| mem-a.16 | 16 | $0.480 |
3. GPU Tier
Information
NERC also supports the most demanding workloads including Artificial Intelligence (AI), Machine Learning (ML) training and Deep Learning modeling, simulation, data analytics, data visualization, distributed databases, and more. For such demanding workloads, the NERC’s GPU-based distributed computing flavor is recommended, which is integrated into a specialized hardware such as GPUs that produce unprecedented performance boosts for technical computing workloads.
There are two flavors within the GPU tier, one featuring older NVidia K80s and the newer NVidia A100s technology.
The "gpu-k80" flavor is provided from Supermicro (2x Intel E5-2620 v3, 24 core,
128GB memory, 2x NVidia K80s) servers. The base service unit is 25% of a whole
server, so 1 SU provides 6 vCPU, 32 GB memory, 1 NVidia K80 at a rate of
$0.190 / hr of wall time. Multiples of the gpu-k80 SU are available with
20 GB root disk.
| Flavor | SUs | Cost / hr |
|---|---|---|
| gpu-k80.1 | 1 | $0.190 |
| gpu-k80.2 | 2 | $0.380 |
| gpu-k80.4 | 4 | $0.760 |
The gpu-k80 flavor will be available soon.
We are still working on setting up the hardware required to enable the cost-effective "gpu-k80" flavor. We will let you know once it is ready and available for your general use.
The "gpu-a100" flavor is provided from Lenovo SR670 (2x Intel 8268 2.9 GHz, 48
core, 384 GB memory, 4x NVidia A100) servers. These latest GPUs deliver
industry-leading high throughput and low latency networking. The base service unit
is 25% of a whole server, so 1 SU provides 12 vCPU, 96 GB memory, 1 NVidia A100
at a rate of $0.537 / hr of wall time. Multiples of the gpu-a100 SU are available
with 20 GB root disk.
| Flavor | SUs | Cost / hr |
|---|---|---|
| gpu-a100.1 | 1 | $0.537 |
| gpu-a100.2 | 2 | $1.074 |
| gpu-a100.4 | 4 | $2.148 |
Pro Tip
Choose a flavor for your instance that suits your requirements, use-cases, and budget when launching a VM.