Easy ReviewCLF-C02
Cloud Technology and Services5

Batch

Run batch computing jobs at any scale without managing batch software or servers.

Este conteúdo de estudo ainda não foi traduzido, então aparece em inglês. Ajude a traduzir

AWS Batch runs batch computing workloads, provisioning compute resources as jobs are submitted.

Managed batch computing

AWS Batch runs batch jobs: units of work, such as simulations or data processing, that run to completion without user interaction. You submit jobs; Batch finds and provisions the compute and spreads the work.

How it works

  1. Define the job and the compute environment it may use.
  2. Submit jobs to a queue.
  3. Batch provisions compute, runs the jobs, and scales back when they finish.

Exam cues

  • If the question says run large numbers of batch jobs without managing servers or schedulers, think AWS Batch.

Remember

  • Fully managed; provisions compute per job demand.

Trap: Batch is for jobs that run to completion. Event-driven functions that respond in milliseconds fit AWS Lambda better.

Da documentação da AWS (7)
  • AWS Batch runs batch computing workloads on AWS and removes the heavy lifting of configuring and managing the infrastructure, like traditional batch computing software.
  • AWS Batch provisions compute resources in response to submitted jobs and optimizes workload distribution by quantity and scale, to remove capacity constraints and reduce compute costs.
  • AWS Batch is fully managed, so there is no batch computing software to install or manage.
  • AWS Batch runs containerized jobs on Amazon ECS or Amazon EKS and scales capacity on EC2 instances (including Spot), Fargate or ECS Managed Instances.
  • Batch multi-node parallel jobs (gang scheduling) run one job across several EC2 instances, with a main node that starts first, for tightly coupled HPC or distributed GPU training.
  • AWS ParallelCluster is an open source tool that deploys HPC clusters with compute, a scheduler such as Slurm and a shared file system; AWS now points new HPC work to the managed AWS Parallel Computing Service.
  • A Lambda function's timeout defaults to 3 seconds and can be set in 1-second steps up to 1,800 seconds (30 minutes).

Flashcards

toque para virar
Resposta

·

Todos os 5 flashcards em lista

Scientists submit thousands of compute jobs that must run on as many servers as needed and then stop. Which AWS service manages this?

AWS Batch, which provisions compute resources for submitted jobs and scales with the workload.

Does a team need to install a batch scheduler to use AWS Batch?

No. AWS Batch is fully managed; there is no batch software to install or manage.

How does AWS Batch reduce compute costs for job workloads?

It provisions resources only for submitted jobs and distributes the work by quantity and scale, so no idle fleet waits for jobs.

A nightly image job runs for three hours and needs a custom runtime. Lambda or AWS Batch?

AWS Batch. Lambda stops at 30 minutes; Batch runs any container image as long as needed on EC2, Spot or Fargate capacity.

Which AWS service manages infrastructure for batch computing workloads?

AWS Batch runs batch jobs and manages the underlying infrastructure.

Você concluiu todos os tópicos da CLF-C02!

Seu badge ficou dourado. Próximo passo: agendar a prova na AWS. Depois de passar, adicione-se ao Quem passou.

Compartilhar

Conte que você concluiu esta certificação

Concluí todos os tópicos da AWS Certified Cloud Practitioner (CLF-C02) no Easy Review. Venha estudar também!