samuelrince
commented
7 months ago Example of a p4d.24xlarge (GPU):
Instance:
- id: p4d.24xlarge
- vcpu: 96
- memory: 1152
- storage_type: ssd
- storage_units: 8
- storage_capacity: 1000
- gpu: 8
- platform: p4d.24xlargePlatform (archetype):
- id: p4d.24xlarge
- manufacturer: AWS
- CASE.case_type: rack
- CPU.units: 2
- CPU.name: Intel Xeon Platinum 8275CL
- RAM.units: 36
- RAM.capacity: 32
- SSD.units: 8
- SSD.capacity: 1000
- HDD.units: 0
- HDD.capacity: 0
- GPU.units: 8
- GPU.name: NVIDIA A100
- GPU.memory: 40
- GPU.connector_type: sxm
- POWER_SUPPLY.units: 2
- POWER_SUPPLY: 2.99;1;5
- USAGE.time_workload: 50;0;100
- USAGE.use_time_ratio: 1
- USAGE.hours_life_time: 35040
- USAGE.other_consumption_ratio: 0.33;0.2;0.6
- USAGE.overcommitted: ??? (true/false or ratio)
- warnings: ...
JacobValdemar
da-ekchajzer
Problem
We want to make the process of adding cloud instances as simple as possible, while:
Resulting in describing both instance characteristics and platform (or bare metal) characteristics.
As of today, both are stored under the same CSV file (cloud archetypes) and can be confusing to interact with. First, it leads to duplicated data about some components (especially CPU with cpu_specs.csv). Second, the contributor needs to understand complex concepts to be able to make a new submission (e.g. difference between vcpu, platform_vcpu, CPU.core_units * CPU.units and USAGE.instance_per_server based on vcpu counts).
Solution
We propose a new way to add cloud instances that should clarify this process. We will separate the concept of a cloud instance and platform (or bare-metal server).
A cloud instance will be described with very few fields that are close to the description provided by cloud providers.
Example of a
c5.2xlarge(in newaws.csv):The platform defined here is
c5.18xlargeis another cloud instance AND a server archetype defined as follows:Cloud instance (also in new
aws.csv):Platform (server archetype):
In this description, the embodied impacts of the cloud instance can be derided from this operation:
$$ Instance{embodied} = \frac{instance.vcpu}{platform.vcpu} * Platform{embodied} $$
We will need to introduce the notion of "vcpu" in
CPUmodeling so that we can take into account the number of "threads" or "virtual cores" in hyper-threading scenarios. Like following:$$ \text{platform.vcpu} = \text{Platform.CPU.units} * \text{Platform.CPU.vcpu} $$
OR:
$$ \text{platform.vcpu} = \text{Platform.CPU.units} \text{Platform.CPU.core\_units} \text{(\# virtual cores per core)} $$
Happy to hear about your feedback @da-ekchajzer. I will detail other examples below.