Skip to main content
Back to challenges
Challenge

Optimize GPU Workload Allocation using LlamaIndex RAG Infrastructure

Cloud software providers facing variable GPU pricing need intelligent routing to optimize infrastructure costs while meeting SLA targets. Build an advanced RAG and agent workflow using LlamaIndex to query cluster performance metrics and dynamically match workloads to low-cost compute clusters. Achieve a 25% cost reduction without breach of SLA.

Business OperationsHosted by Vera
Sign in free to start
Challenge brief

What you are building

The core problem, expected build, and operating context for this challenge.

Build an AI compute scheduler using LlamaIndex RAG agents to match AI workloads to optimal GPU nodes based on cost and SLA constraints.

Delivery guide

How work is evaluated

Evaluation

Evaluates workload allocation decisions for cost savings and SLA adherence.

Datasets

Challenge datasets

Public datasets, plus the private datasets you have access to.

Evaluation rubric

How submissions are scored

These dimensions define what the evaluator checks and which criteria separate a passable run from a strong one.

Dimensions
2 scoring checks
Binary
2 pass or fail dimensions
Ordinal
0 scaled dimensions
Dimension 1

sla_compliance_check

Ensures selected node meets max_latency_sec constraint

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Dimension 2

cost_efficiency

Ratio of optimal cost achieved vs naive node selection • target: 0.85 • range: 0-1

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Learning goals

What you should walk away with

  • Implement advanced RAG indexing over cluster performance manuals and pricing APIs with LlamaIndex

  • Build LlamaIndex QueryEngine tools for autonomous agent routing decisions

  • Synthesize real-time telemetry on GPU utilization and hourly spot instance rates

  • Produce cost-optimized job scheduling outputs with strict SLA constraints

Resources and assets

Reference links and supporting material

Dataset notes

Cloud GPU pricing logs, cluster utilization metrics, and hardware performance specs.

Challenge setup

Evaluates workload allocation decisions for cost savings and SLA adherence.

Challenge input

JSON object with workload specs and current node pricing

LlamaIndex

Offers structured data connectors and context retrieval agents.

Evaluated output

JSON object with selected node, estimated cost, and SLA status

Evaluation checks
  • Ensures selected node meets max_latency_sec constraint
  • Ratio of optimal cost achieved vs naive node selection • target: 0.85 • range: 0-1
Configured evaluation evidence
  • Benchmark: Cloud Compute Optimization Benchmark
  • Cost Efficiency target: 0.85
  • 1 public reference case
Configured runtime
  • Python execution harness
  • Python sandbox (unavailable on Versalist)

Summary shows the first task, up to three tools, and up to three evidence items. Open the technical recipe to view all configured items.

View technical recipe

Configured tools

Action Space
  • Llama Index · Required
  • Hyperbolic · Optional
  • Ellipsis · Optional

Evaluation contract

  • sla_compliance_check · Weight 1
  • cost_efficiency · Weight 1

Metrics

  • cost_efficiency · Target 0.85

Benchmarks

  • Cloud Compute Optimization Benchmark
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON object with workload specs and current node pricing

    Output: JSON object with selected node, estimated cost, and SLA status

    Execution harness: Python

Evaluation tools

  • LlamaIndex · RAG Framework

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Optimize GPU Workload Allocation using LlamaIndex RAG Infrastructure