벤더 질문:
"what is the main goal of GPU cluster? Could GPU... both training
[and] inference. So GPU, they don't have an answer, like, whether
it's for training or inference"
현황:
벤더 설명:
"I wanted to understand better the needs or the plan for the
deployment... so we can size the AI store recommendation"
목표:
1. Training vs Inference 요구사항 파악
2. 스토리지 용량 산정
3. 네트워크 처리량 계산
4. 하드웨어 권장사항
벤더 설명:
"VT content has eight GPUs... Each of these GPUs is connected
into its own [NIC]... And then there's two CPUs in this platform,
and then there's two [NICs]... these ones are 400 [Gbps]"
DGX 노드 구성:
┌─────────────────────────────────────────┐
│ NVIDIA DGX Node │
├─────────────────────────────────────────┤
│ 8x GPUs (H100/A100/etc) │
│ ↓ │
│ 2x CPUs │
│ ↓ │
│ 2x 400Gbps NICs (ConnectX or Bluefield)│
└─────────────────────────────────────────┘
최대 처리량: 800Gbps (2 x 400Gbps)
벤더 설명:
"Nvidia has this super[POD] specification... Nvidia says, take
one of these and put 72 together, right? 72 of these, which
[is] 76 GPUs [total]."
SuperPOD 구성:
벤더 설명:
"when you're training, [you can] pull between hundreds megabytes
a second to 500 megabytes per GPU"
처리량:
벤더 계산:
"this gives you... a theoretical [throughput] of 64 gigabytes a
second to 288 gigabytes [per SuperPOD]"
SuperPOD 처리량:
벤더 계산:
"for me[MinIO], this will be between four to six [MinIO] nodes...
assuming MinIO has [dual] 400 gigabit [NICs]"
MinIO 클러스터:
72 DGX (SuperPOD) → 4~6 MinIO 노드
- MinIO 노드당: Dual 400Gbps (800Gbps total)
- 처리량: 충분 (64~288 GB/s 대비)
벤더 강조:
"training is predictable. You know how much data you [will be]
pulling, how long you [will be] training for, and what's [the]
expected throughput"
특성:
벤더 설명:
"on [inference], the math changes because of the models. So...
the size of the model... and the number of [concurrent] users changes."
변수:
벤더 설명:
"Let's say Qwen 2.5... one of the largest ones is roughly 610
gigabytes of weights... you need to load into memory."
모델 로딩:
벤더 설명:
"the process of inference... is split into two parts... they
call [prefill] and [decode]"
2단계 처리:
1. Prefill (전처리):
┌──────────────┐
│ GPU Node 1 │
│ (Prefill) │
│ │
│ Input tokens │
│ → LLM │
│ → KV Cache │
└──────────────┘
2. Decode (생성):
┌──────────────┐
│ GPU Node 2 │
│ (Decode) │
│ │
│ KV Cache → │
│ → Generate │
│ → Output │
└──────────────┘
벤더 설명:
"when [you prefill] the tokens into the LLM, it generates
something called KB cache... that needs to be moved over to
the GPU node for decode"
KV Cache 저장 위치:
1. GPU to GPU: 가장 빠름
2. GPU to CPU: 중간
3. GPU to Local NVMe: 느림
4. GPU to Object Storage: 가장 느림
처리량 (Qwen 2.5 기준):
"requires... a throughput per... model like such as this one,
between 18 gigabytes a second to 70 gigabytes"
벤더 설명:
"if you're doing [distributed] inference, and you do this because
you want to be able to not have to recompute in conversations
every time"
목적:
벤더 언급:
"If you're going to do smaller models... this math is way simpler"
Computer Vision 모델:
고객사 설명:
"as of today, inference... they're building 300 agents just for
today. So like, they will have a lot of requests... for inferences."
현황:
계획:
"which will be slowly moving to [MinIO] at the end [of] today"
고객사 설명:
"fault detection... for... semiconductor production... trying to
figure out... what's going on... to prevent you from having a fault"
용도:
고객사 설명:
"their plan is to... run this [fault] detection, FDC, live stream,
real time training and analysis going forward next year."
타임라인:
고객사 설명:
"the data, or the training itself is already five petabytes...
They are starting from five petabytes"
현재:
"they have another five petabyte of data... which is stored...
outside of [AIStore]... like somewhere else in legacy."
위치:
"they also have archival data, 15 petabyte, which is also stored
outside of [AIStore]."
총계:
고객사 설명:
"if you just think about 2027 they will have 10 petabytes of data
running in [MinIO] cluster with GPU. Five petabyte out of 10
petabyte will be [the] data for training."
2027년 목표:
벤더 확인:
"if [the] inference is done around computer vision models.
Computer Vision models are way smaller and [can do] inference
in a single box"
고객사 확인:
"computer vision. It will be 100%. It will be 100%"
특성:
벤더 설명:
"Computer Vision... you still need to stream but the amount of
data that you stream still 300 to 500 megabytes per [GPU]."
Per GPU:
벤더 권장:
"a simple [MinIO] cluster, roughly 40[to]45 nodes will be more
than enough... especially [with] 400 gigabit"
권장 사양:
벤더 설명:
"400 gigabit, [the] speed of the NVMe is shared... it can be
doubled on the same nodes, If you go to dual socket, as opposed
to single socket"
권장:
고객사 계획:
"what they're planning for [Yongin] cluster is they will have
15 [drives], 30 terabyte NVMe, [per node]."
문제:
"the realistic issue is, there's no 30 terabyte NVMe. They're
working on it."
Pool 0 계획:
고객사 우려:
"getting the second pool at the same size will be almost impossible
because of budgets and protocols"
계획:
벤더 확인:
"the only requirement for subsequent pools is that the parity
value is similar... the size of the pool could be... larger
or it could be smaller."
성능:
고객사 설명:
"they're like, building this AI factory... [in] Yongin. It's a...
place in Korea"
목적:
고객사 설명:
"we don't have electricity... They will need a [better]...
infrastructure... plants and everything. So they're considering
some other place."
문제:
고객사 계획:
"Eventually, at the end of... maybe... three years [forward],
they will have this... huge cluster... where they have like
3000 [storage] nodes... and they will have... 5000 nodes...
or maybe... starting from... 1000 nodes of GPU farm"
목표 (2029년):
고객사 확인:
"they will also use [InfiniBand] as back end... for [GPU],
that's confirmed."
벤더 언급:
"the hardware that we recommended last time, being expensive is
one thing, but the second thing is... they can't even get it in
the market"
문제:
벤더 제안:
"my [customer] Texas Instruments... went with Cisco and Lenovo
because they had the seven and a half terabyte NVMe drives in
stock, because they bought [them] earlier."
권장:
벤더 권장:
"roughly 40[to]45 nodes will be more than enough... especially
[with] 400 gigabit"
사양:
5~10PB 목표:
DGX 노드:
MinIO 노드:
Per GPU:
Per DGX (8 GPUs):
100 DGX:
MinIO (40 nodes):
벤더 설명:
"kb cache. Sometimes you don't even need it. People come ask
one question and they never come back. So restoring KV cache
is not something so needed."
Computer Vision:
대규모 모델만 해당:
Computer Vision:
GPU 워크플로우 문서
하드웨어 권장사항
사이징 가이드
하드웨어 조달
인프라 준비
워크로드 명확화
Pilot 구축
데이터 마이그레이션
FDC Production
확장
Computer Vision Training:
예상:
하드웨어:
운영:
3년 비전:
문서 버전: 1.0
최종 수정일: 2026년 2월 4일
다음 리뷰: Pilot 완료 후 (2026년 Q2)
"training is predictable. You know how much data you [will be]
pulling... but on [inference], the math changes because of the models"
"if [the] inference is done around computer vision models.
Computer Vision models are way smaller and [can do] inference
in a single box"
"the data, or the training itself is already five petabytes...
They are starting from five petabytes"
"a simple [MinIO] cluster, roughly 40[to]45 nodes will be more
than enough... especially [with] 400 gigabit"
"they will have this... huge cluster... 3000 [storage] nodes...
and... 1000 nodes of GPU farm"