Add m8id instance family (Granite Rapids + local NVMe)#291
Open
JugadK wants to merge 2 commits into
Open
Conversation
JugadK
force-pushed
the
add-m8id
branch
4 times, most recently
from
dea7625 to
1632117
Compare
m8id = m8i CPU + local NVMe. Mirrors m8i (GRANITE_RAPIDS_IPC, 3.9 GHz all-core) and adds the local-NVMe ephemeral profile: - instance_families.py: "m8id" entry with io_latency_curve "8th-gen-ssd". - auto_shape.py: "8th-gen-ssd" latency curve from fio (4k randread, QD1, 60s), median across 49 m8id.4xlarge (three runs): low/mid/high = p5/p50/p95 = 0.070/0.093/0.105 ms. Per-instance p50 is bimodal (~76us / ~93us, ~35%/65% split reproduced across all three runs), so the median is firmly ~93us. minimum_value 0.03 = realistic floor (clat-min ~0.5us is a cached-I/O artifact); maximum_value 1.2 = conservative tail bound. - auto_shape.py: aws_iops_per_gib "8id" = (303.5, 151.8). m8id's published instance-store IOPS match the m6id/m7gd per-GiB rate exactly (m8id.96xlarge = 6,439,992 read / 3,220,008 write over 6x3800 GB = 22,800 GB).
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Adds the m8id instance family: Intel Xeon 6 (Granite Rapids) CPU — identical to m8i — plus local NVMe ephemeral storage.
instance_families.py: newm8identry. Shares m8i's CPU (GRANITE_RAPIDS_IPC, 3.9 GHz single/all-core turbo); uses a new8th-gen-ssdio-latency curve.auto_shape.py: new8th-gen-ssdlatency curve, measured via fio (4k randread, QD1, 60s; median across 10× m8id.4xlarge): p5/p50/p95 = 0.074 / 0.077 / 0.094 ms.minimum_valueis a realistic 0.03 ms floor (the fio clat-min ~0.5 µs is a cached-I/O artifact, not the device floor).TODO before un-drafting
hardware/profiles/shapes/aws/auto_m8id.json(python -m service_capacity_modeling.tools.generate_missing --execute; needs EC2describe-instance-typescreds).tox -e pre-commitandtox -e py312.