Scientific Workflow Engine Selection Matrix
Compare and score distributed computational pipeline runners for high-throughput research workloads.
Use this template when evaluating workflow orchestration engines for reproducible scientific analysis, lab data pipelines, or high-throughput batch processing. It allows infrastructure teams to contrast tool capabilities against strict lab requirements.
Role: Lead Research Software Engineer specializing in high-performance computing infrastructure and scientific workflow automation.
Context
- Scientific domain: {{research_discipline}}
- Pipeline computation profile: {{workload_characteristics}}
- Hosting infrastructure: {{compute_environment}}
- Candidate workflow tools: {{orchestration_candidates}}
- Data reproducibility baseline: {{reproducibility_standard}}
Task
Generate a multi-criteria decision matrix evaluating {{orchestration_candidates}} to identify the optimal pipeline orchestrator for {{research_discipline}} workloads on {{compute_environment}}.
Method
- Translate {{workload_characteristics}} into measurable infrastructural dimensions (e.g., I/O throughput, containerization, GPU dispatch).
- Establish weighted scoring criteria based on portability, provenance tracking, fault tolerance, and {{reproducibility_standard}}.
- Evaluate each tool in {{orchestration_candidates}} against bare-metal and cloud capabilities of {{compute_environment}}.
- Score each candidate across criteria on a 1-5 standardized scale.
- Compute normalized aggregate scores showing clear technical differentiation.
- Highlight critical operational failure modes and community maturity for top contenders.
- Formulate a final tool recommendation backed by tradeoff rationale.
Constraints
- MUST format the core evaluation as a clean Markdown weighted scorecard matrix.
- MUST NOT recommend tools lacking native support for {{reproducibility_standard}}.
- Scoring justifications MUST cite specific architectural capabilities rather than general popularity.
- Limit overall narrative to concise rationale preceding and following the matrix.
Output format
- Evaluation Criteria & Weighting Key (bullet list with weights summing to 100%)
- Orchestrator Comparison Matrix (Markdown table with columns: Evaluation Dimension, Weight, Candidate Scores [1-5 for each candidate], Weighted Totals)
- Trade-off Analysis & Recommendation (2-3 concise summary paragraphs)
Self-review
- Does the matrix clearly reflect the constraints of {{compute_environment}}?
- Are all candidates from {{orchestration_candidates}} scored across every criterion?
- Does the recommended engine demonstrably handle {{workload_characteristics}}?
Explicit role, a named task, and discrete steps the model can follow.
Background, inputs and variables the model needs before it starts.
Hard boundaries — what the model must and must not do.
A named, field-level shape for the response.
Ordered work items that force analysis before an answer.
Length and structure that travel across frontier models.
Signal density — instruction weight without padding.
Documented variables so the scaffold adapts to new inputs.
Quality bar, assumptions and behaviour when inputs are thin.
How much real usage the template has behind it.