Scale AI

Module 4 · Prepare Your Workload for Compute Access
⏱️ Est. 50–60 min

Scale AI · Technical Core

Prepare Your AI Workload for Compute Access

Learn how data moves through its lifecycle, which infrastructure constraints shape an AI workload, how hosting choices differ and what working with high-performance computing (HPC) means. The goal is to prepare a credible request for future compute access—not to design or administer a production cluster.

Compute access is more than receiving a GPU

Useful access depends on three descriptions that must fit together. A powerful machine cannot compensate for an undefined workload, inaccessible data or an unreproducible software environment.

Workload · what must run

The task, job type, scale, frequency, target result and acceptable completion time.

Data path · what must move

Sources, preparation, permissions, volume, storage, transfer, outputs and retention.

Operating boundary · how it may run

Users, access controls, software, logging, quotas, support, ownership and cleanup.

Example · monthly speech-model evaluation: Workload: run one bounded batch evaluation of a fixed model against 8,000 approved audio clips and return accuracy plus error classes within four hours. Data path: create a versioned, minimised snapshot in the research archive, transfer it to temporary compute storage, verify the copy, return only metrics and logs, then delete the staged files after seven days. Operating boundary: two named researchers use an approved container under a fixed quota; every run is logged, no model is deployed and the project owner accepts the result.

What this module prepares you to do

Describe the whole workloadConnect the job, its data path and its operating boundary.
Challenge a resource requestAsk what evidence supports the hardware, hosting and runtime estimate.
Prepare the access conversationLeave with a Compute Readiness Brief that an infrastructure team can review.
1

Start with the workload, not the hardware

Translate the use case into work the infrastructure must perform before naming a cloud, cluster or accelerator.

What is a workload?

A workload is a bounded unit of computational work together with its input, output and operating requirements. “Train a model” is still too vague. Sizing begins with job type, data volume, frequency, acceptable duration, memory pressure, parallelism and the evidence the run must produce.

Compute is not one thing. A central processing unit (CPU) is flexible general-purpose compute. A graphics processing unit (GPU) performs many similar numerical operations in parallel and can accelerate suitable AI workloads. More GPUs do not guarantee a proportionally faster run: data loading, memory, software and communication can become the bottleneck.

Workload shapePrimary concernTypical first question
Interactive inferenceResponse latency, availability and predictable concurrencyHow quickly must one user receive a result?
Batch inferenceThroughput, scheduling and cost per processed itemCan the work run hourly, nightly or on demand?
Fine-tuningGPU memory, dataset size, runtime and checkpointsWhich model, precision and training method will be used?
Large training or simulationParallelism, interconnect, storage throughput and reproducibilityCan the work use multiple nodes efficiently?

Right-sizing principle

Request the smallest resource profile that can answer the next decision. Start with a short representative run, observe the bottleneck and revise the estimate before scaling the job.

Activity · Which description is compute-ready?

All three teams want GPU access. Select the description that gives an infrastructure specialist enough information to start a sizing conversation.

✓ Workload description recorded
2

Trace the data lifecycle before moving data

Compute readiness begins at the source and continues after the job finishes.

One lifecycle, several gates

The data lifecycle describes how raw signals are collected, prepared, governed, made available to compute, transformed into results and eventually retained or removed. Every stage needs an owner and a minimum control. If a source definition, permission or quality rule is unclear, faster compute only produces unreliable results sooner.

White · data movementDescribes where data originates, enters the workflow or is staged.
Blue · compute activityMarks the stage where the approved snapshot is processed.
Green · control gateShows where quality, access or retention must be checked before proceeding.

Three controls travel through the entire flow

The boxes below are not extra stages. They are questions that connect several numbered stages. A compute result is trustworthy only when the team can follow these controls from the source to the retained output.

Stages 1–3 and 6

Quality

Define meaning and representative cases at the source, validate during ingestion, apply the quality gate during preparation and check whether compute results reveal new data problems.

Stages 1, 3, 4, 6 and 7

Lineage and version

Data lineage is the traceable record of where data came from, how it was transformed and versioned, where it moved and which outputs it produced. Link the source snapshot, preparation logic, labels, software and output so another authorised person can explain and reproduce the result.

All stages · especially 4, 5 and 7

Security and retention

Apply permitted use and least-privilege access throughout, protect the staged copy and decide which inputs, checkpoints, logs and outputs are returned, archived or deleted.

Optional deep dive · Storage close to compute

Large jobs are often limited by how quickly data reaches processors, not by processor speed alone. Repeatedly reading many small files across a slow connection can leave expensive GPUs idle even though the resource request looked sufficient on paper. This is why capacity—how much data fits—and throughput—how quickly it can be read or written—must be considered separately.

Authoritative storage and compute storage have different roles. The governed source remains the system of record. For a bounded run, the team may create an approved, versioned snapshot and stage that copy in storage designed for the compute environment. The staged copy is temporary working material, not a new uncontrolled master dataset.

File layout matters. Millions of very small files can create more overhead than a smaller number of well-organised files with the same total size. A representative transfer and loading test can reveal whether packaging, indexing or a different data format is needed before costly processors are reserved.

Integrity must be demonstrated. Record the snapshot identifier, file count and expected size; use checksums or an equivalent validation to show that the copy arrived unchanged. Failed, missing or quarantined records must be visible rather than silently omitted from the experiment.

Security follows the copy. Apply least-privilege access in the temporary environment, encrypt transfers where required, avoid unnecessary fields and define whether checkpoints or logs could reproduce sensitive information. Access to compute does not automatically grant access to every dataset.

Plan the exit before the transfer. Decide which metrics, checkpoints or logs return to the project workspace, who accepts them, when temporary data is deleted and what evidence proves cleanup. Include storage duration, transfer volume and cleanup effort in the cost and operating estimate.

Practical readiness check: before requesting the full run, transfer and load a representative sample. Record elapsed time, effective throughput, failed files, temporary storage used and the owner who approves the result. That small test turns a data-path assumption into sizing evidence.

Activity · Which action makes this data path compute-ready?

A research team wants to evaluate a model on case documents. Use of a redacted snapshot is approved, but no snapshot has been versioned, its transfer has not been checked and nobody owns deletion after the run. What should happen before requesting more compute?

✓ Lifecycle gate identified
3

Understand the infrastructure building blocks

Infrastructure is a delivery system: processors matter, but so do memory, storage, networking, software and operational controls.

Compute

CPU, GPU and accelerators

Execute the work. Suitability depends on the algorithm, supported software, precision and degree of parallelism.

Memory

Working space

Holds model parameters, activations and batches during a run. A job that does not fit cannot use the processor effectively.

Storage

Data and artefacts

Holds datasets, containers, checkpoints, logs and outputs. Capacity and throughput are different requirements.

Network

Movement and coordination

Transfers data and connects nodes. Distributed jobs may depend on low-latency, high-bandwidth interconnects.

Software environment

Reproducible execution

Drivers, frameworks, libraries, containers and versions must match the hardware and the tested workflow.

Control plane

Safe shared use

Identity, permissions, scheduling, quotas, logging, monitoring, support and cost attribution make resources governable.

The bottleneck moves

A slow job does not automatically need more GPUs. It may be waiting for data, exceeding memory, using software that cannot parallelise, communicating too much between nodes or requesting resources it leaves idle.

Activity · Match each symptom to the first infrastructure question

✓ Infrastructure map recorded
4

Choose a hosting direction from constraints

Hosting determines where data and models run, who operates them and how much control, speed and responsibility the organisation accepts.

Managed service / API

Fastest start

Provider operates the service. Strong for learning and variable use; assess data movement, service limits, dependency and unit cost.

Cloud infrastructure

Flexible capacity

Rent configurable compute and storage. Offers elasticity but still requires identity, security, cost control and operations.

Self-hosted / on-prem

More direct control

Organisation operates hardware or private infrastructure. Control comes with staffing, maintenance, utilisation and lifecycle responsibility.

Shared HPC

Specialised batch compute

Access powerful shared resources through projects, quotas and a scheduler. Suits bounded parallel jobs more than always-on services.

Decision dimensionQuestions to answer
Data and sovereigntyWhere may inputs, logs, checkpoints and outputs be stored or processed?
Workload shapeInteractive or batch? Continuous demand or occasional burst? One node or many?
OperationsWho patches, monitors, supports, secures and updates the environment?
EconomicsWhat drives variable cost, idle capacity, transfer cost, licences and staff effort?
OptionalityCan data, containers, artefacts and evaluation evidence move to another environment?

Hybrid is a design, not indecision

A hybrid approach can keep sensitive source data in an existing controlled environment, use temporary external compute for an approved snapshot and return only defined outputs. It earns its complexity only when a real constraint requires more than one environment.

Activity · Choose a hosting direction for a constrained workload

Scenario: sensitive source records must remain in the existing controlled environment. An approved, minimised snapshot may be transferred for occasional GPU-intensive batch training, and only checkpoints and evaluation outputs should return.

✓ Hosting direction recorded
5

Understand how shared HPC changes the workflow

High-performance computing is powerful shared infrastructure. Users normally describe and submit bounded jobs rather than treating a machine like a permanently available laptop.

HPC in plain language

High-performance computing (HPC) combines powerful processors, large memory, fast storage and high-speed networking so demanding jobs can run efficiently. A cluster contains multiple computers, called nodes. A scheduler allocates shared resources according to job requests, priorities and availability.

A submitted job may wait in a queue before it runs. This is normal resource sharing, not a failure. A good request states the processors or GPUs, memory, expected runtime and software environment it needs. Over-requesting wastes scarce capacity and can increase queue time; under-requesting can make the job fail.

Node

One computer in a cluster, with its own processors and memory.

Partition / queue

A resource pool or policy lane for particular job types and limits.

Scheduler

The service that decides when and where submitted jobs run.

Batch job

A non-interactive task defined in advance and executed when resources become available.

Checkpoint

A saved intermediate state that can protect long runs from total restart.

Parallel efficiency

How much useful speed is gained when more processors or nodes are added.

White · project-team workPreparation, resource description and monitoring remain the user's responsibility.
Blue · platform hand-offThe scheduler receives the bounded job and places it in shared infrastructure.
Green · control gateApproval before the run and accountable closure afterwards protect shared use.
Data readyApproved snapshot, size, format, transfer and retention known.
Environment readyCode, dependencies, container or modules and versions documented.
Job readyResource estimate, runtime, checkpoints, logs and success criteria defined.
People readyUser identities, owner, support route and result reviewer named.

Activity · Which workload is a plausible HPC candidate?

✓ HPC fit recorded
6

Prepare a Compute Readiness Brief

Give an access or infrastructure team enough context to challenge the workload, recommend a starting resource profile and identify prerequisites.

A brief is a sizing hypothesis, not a purchase order

State what is known, label estimates and begin with a representative trial. The receiving team may recommend a smaller resource, different job shape, alternative hosting path or more data preparation before access.

Purpose and success

Decision supported, workload outcome, evaluation baseline and acceptable runtime.

Technical envelope

Job type, data volume, resources, software, storage, transfer and expected outputs.

Access and operation

Users, permissions, classification, owner, support, quota, logging, retention and cleanup.

Activity · Generate your Compute Readiness Brief

Use a fictional or sanitised workload. Do not enter dataset locations, credentials, personal data or sensitive system details.

Complete at least seven fields to generate your brief.
✓ Compute Readiness Brief recorded

Scale AI technical summary

Six takeaways before requesting compute

Describe the workload, outcome and acceptable runtime before naming hardware.
Trace data from source through staging, compute, outputs, retention and feedback.
Treat compute, memory, storage, network, software and controls as one delivery system.
Choose hosting from data, workload, operational, economic and optionality constraints.
Expect shared HPC to use bounded jobs, resource requests, queues, checkpoints and cleanup.
Submit a sizing hypothesis with explicit evidence, owners, prerequisites and a small first trial.

Leadership questions

Which lifecycle gate can block the workload? What is the likely bottleneck? Why is this hosting direction proportionate? Can the team reproduce the environment? What will the first compute run prove, and who decides what happens next?