<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body>
<div class="" style="word-wrap:break-word"><span class="" style="font-size:14.666666984558105px">This is an announcement of Jin Jin Zhao's MS Presentation</span><br class="" style="font-size:14.666666984558105px">
<span class="" style="font-size:14.666666984558105px">===============================================</span><br class="" style="font-size:14.666666984558105px">
<span class="" style="font-size:14.666666984558105px">Candidate: Jin Jin Zhao</span><br class="" style="font-size:14.666666984558105px">
<br class="" style="font-size:14.666666984558105px">
<span class="" style="font-size:14.666666984558105px">Date: Wednesday, May 25, 2022</span><br class="" style="font-size:14.666666984558105px">
<br class="" style="font-size:14.666666984558105px">
<span class="" style="font-size:14.666666984558105px">Time: 3 pm CST</span><br class="" style="font-size:14.666666984558105px">
<br class="" style="font-size:14.666666984558105px">
<span class="" style="font-size:14.666666984558105px">Location: JCL 298</span><br class="" style="font-size:14.666666984558105px">
<br class="" style="font-size:14.666666984558105px">
<span class="" style="font-size:14.666666984558105px">M.S. Paper Title: AUTOMATED PROVENANCE CAPTURE IN ARRAY-PROGRAMMING FRAMEWORKS</span><br class="" style="font-size:14.666666984558105px">
<br class="" style="font-size:14.666666984558105px">
<span class="" style="font-size:14.666666984558105px">Abstract: This paper presents DSLog, a system that efficiently capture and represent fine-grained data</span><br class="" style="font-size:14.666666984558105px">
<span class="" style="font-size:14.666666984558105px">provenance in array-programming frameworks for black box functions. It uses a technique</span><br class="" style="font-size:14.666666984558105px">
<span class="" style="font-size:14.666666984558105px">called annotated execution to capture “physical” provenance, automatically without user</span><br class="" style="font-size:14.666666984558105px">
<span class="" style="font-size:14.666666984558105px">specification. We describe a low-level implementation for arrays up to</span><br class="" style="font-size:14.666666984558105px">
<span class="" style="font-size:14.666666984558105px">100 million (and more) cells. This implementation also improves capture performance up to 34x over a high level</span><br class="" style="font-size:14.666666984558105px">
<span class="" style="font-size:14.666666984558105px">baseline. Additionally, we contribute a new compression algorithm, named ProvRC, that compresses such</span><br class="" style="font-size:14.666666984558105px">
<span class="" style="font-size:14.666666984558105px">relations. We show that the ProvRC results in a significant storage reduction over functions</span><br class="" style="font-size:14.666666984558105px">
<span class="" style="font-size:14.666666984558105px">with simple spatial regularity, beating alternative baselines by many orders of magnitude.</span><br class="" style="font-size:14.666666984558105px">
<span class="" style="font-size:14.666666984558105px">Finally, we present the concepts of dimensional and generalized views over these compressed</span><br class="" style="font-size:14.666666984558105px">
<span class="" style="font-size:14.666666984558105px">relational representation, which allows DSLog to recognize previously seen function (with</span><br class="" style="font-size:14.666666984558105px">
<span class="" style="font-size:14.666666984558105px">only input array dimension information, and no input array information respectively), and</span><br class="" style="font-size:14.666666984558105px">
<span class="" style="font-size:14.666666984558105px">re-use pre-existing materialized provenance views. We demonstrate that these views cover</span><br class="" style="font-size:14.666666984558105px">
<span class="" style="font-size:14.666666984558105px">92% and 73% respectively of 136 tested numpy functions, and preliminary results show that</span><br class="" style="font-size:14.666666984558105px">
<span class="" style="font-size:14.666666984558105px">using the views have a marked improvement over pure naive annotated execution.</span><br class="" style="font-size:14.666666984558105px">
<br class="" style="font-size:14.666666984558105px">
<span class="" style="font-size:14.666666984558105px">Advisors: Sanjay Krishnan</span><br class="" style="font-size:14.666666984558105px">
<br class="" style="font-size:14.666666984558105px">
<span class="" style="font-size:14.666666984558105px">Committee Members: Sanjay Krishnan, Raul Castro Fernandez, Blase Ur, and Nick Feamster</span><br class="" style="font-size:14.666666984558105px">
<div class=""><span class="" style="font-size:14.666666984558105px"><br class="">
</span></div>
<div class=""></div>
</div>
<div class="" style="word-wrap:break-word">
<div class=""></div>
<div class=""><span class="" style="font-size:14.666666984558105px"><br class="">
</span></div>
</div>
</body>
</html>