-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathaws.html
More file actions
86 lines (76 loc) · 4.06 KB
/
Copy pathaws.html
File metadata and controls
86 lines (76 loc) · 4.06 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
<title>SAT Competition</title>
<link rel="stylesheet" href="main.css" type="text/css">
<link rel="stylesheet" href="template.css" type="text/css">
<link rel="icon" type="image/x-icon" href="doge2.ico">
<script src="https://www.w3schools.com/lib/w3.js"></script>
<style>a#aws { color:var(--link-color-2); }</style>
</head>
<body>
<div class="main">
<div class="navigation" w3-include-html="navigation.html"></div>
<div class="heading" w3-include-html="heading.html"></div>
<script>w3.includeHTML();</script>
<div class="content">
<h2>Parallel and Cloud Track Submissions</h2>
Please make sure to adhere to the <a href="rules.html">General Rules</a>.<BR><BR>
The Parallel and Cloud tracks will be run on Amazon Web Services. The
competition infrastructure has been rewritten for 2026. Instructions
and the full package are available here:
<a href="https://github.com/aws-samples/aws-batch-comp-infrastructure-sample">
https://github.com/aws-samples/aws-batch-comp-infrastructure-sample</a>.<BR><BR>
The instance type codes are documented in the AWS documentation.<BR><BR>
The new infrastructure is designed to be simple. You submit only two
files: a Dockerfile to build your solver, and a small Python shim
(`solver_cmd.py`) to connect your solver’s inputs and outputs to the
package. A single CLI script (`satcomp.py`) is the entrypoint for
everything: building, local testing, deploying to AWS, running jobs,
and collecting results.<BR><BR>
Key points:<BR>
1. Local-first development. Build and test your solver in local Docker
containers before testing with AWS.<BR>
2. One tool for everything. `satcomp.py build`, `test-local`, `push`,
`start-instances`, `submit`, `collect`, `teardown` — all from one
script.<BR>
3. Same package runs the competition. The package you use to prepare
your solver is the same one used to run the competition.<BR>
4. Lots of examples. The repository includes Dockerfiles and shims for
multiple solvers from previous competitions, so you can get started
quickly.<BR>
5. Qualification testing. Built-in `test` command validates your
solver against submission requirements before you submit.
<!--
<p>
Please make sure to adhere to the <a href="rules.html">General Rules</a>.
</p>
<p>
The Parallel track will be run on Amazon Web Services.
Participants of this track are required to submit their solver via
a GitHub repository. The instructions for submissions are available here:
<a href="https://github.com/aws-samples/aws-batch-comp-infrastructure-sample">
https://github.com/aws-samples/aws-batch-comp-infrastructure-sample</a>.
</p>
<p>
The instance type codes are documented in the <a href="https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/cpu-options-supported-instances-values.html">AWS documentation</a>.
</p>
<p>
The AWS team worked hard to improve the developer experience for participants of the parallel track. As compared with the 2022 infrastructure, the following has been added or modified:<br />
1. Scripts and instructions for local debugging of both parallel and distributed solvers on your workstation/laptop.<br />
2. Automated upload of stdout/stderr and intermediate files for each run to S3 storage so that you can debug runs after completion.<br />
3. A "quickstart" that allows you to build the infrastructure and solver, then run it using only two commands.<br />
4. Default values for most of the required arguments from 2022 to make it much simpler to run commands.<br />
5. A richer input.json format that is used for solver communication that allows passing arguments to solvers,
a timeout, and information about the input format.<br />
6. A change in the expected solver output format in solver_out.json. This year, the satisfiability/unsatisfiability
of the problem is indicated by the return code of the solver (10 for SAT, 20 for UNSAT)<br />
7. Various improvements and bug fixes in the scripts for building/running jobs to provide better observability
of the process.
</p>
-->
</div>
</div>
</body>
</html>