Wail Alkowaileet

About Me

I am a Ph.D. candidate in the Information Systems Group (ISG) at the University of California, Irvine under the supervision of Professor Michale J. Carey. I’m also a research associate at the Center for Complex Engineering Systems at KACST and MIT.

My research targets the storage engines in Big Data Management Systems. In particular, I work on reducing the storage size to accelerate scan-based analytical workloads for document store systems.

Education

PhD, Computer Science

2017 - Present

University of California, Irvine

Topics: Storage and query optimization in LSM-based document store systems.
Supervisor: Michael J. Carey

MSc, Computer Science

2011 - 2013

University of California, Irvine

Thesis: NUMA-aware Multicore Matrix Multiplication.
Supervisor: Isaac D. Scherson

BSc, Computer Science

2011 - 2013

King Saud University

Experience

Committer

2016–Present

Apache AsterixDB

asterixdb.apache.org

Components: Storage and Query Optimization

Research Associate
Research Affiliate

2014–Present

King Abdulaziz City for Science and Technology (KACST)

Massachusetts Institute of Technology (MIT)

cces.kacst.edu.sa

Center for Complex Engineering Systems (CCES) – Institute for Data, Systems and Society (IDSS)
Role: Developing tools capable of harnessing and analyzing large-scale data
Projects: AsterixDB-Spark Connector, CityDynamics, Connected Intelligence Platform, Integrated Transportation Systems, Innovation Space

Associate Software Engineer

2008–2009

Advanced Electronics Company (AEC)

Research and Development Department (R&D)
Role: Developing components that connect electric and water smart-meters to the data collection units
Projects: ADDAD4, Water Smart Meter

Projects

LSM-based Tuple Compaction Framework

We proposed a new mechanism to leverage LSM-lifecycle events to infer the schema and semantically compact self-describing semi-structured records automatically. We also introduced a novel semi-structured record physical format for efficient construction and compaction. Using Apache AsterixDB, we were able (combined with our implementation of page-level compression) to reduce the storage size by 9.8x and improve the query performance by the same factor.
Paper: Extended version in arXiv

PAX for LSM-based Document Store Systems

We are currently Investigating the feasibility Partitioned Attributes Across (PAX) page format for LSM-based document store systems as an alternative to the N-ary Storage Model (NSM or slotted pages). One disadvantage of the PAX page was the ability to do in-place updates. As for LSM indexes, in-place updates are not permitted, which makes PAX page format more appealing.

Awards

Ph.D Schoalrship

2017–2021

Awarded full graduate scholarship from the King Abdulaziz City for Science and Technology.

MSc Schoalrship

2010-2013

Awarded full graduate scholarship from King Abdullah Scholarship Program.

Second Class Honor

2004-2008

Awarded Second Class Honor for high GPA from King Saud University

Publications

[1]

W. Alkowaileet, S. Alsubaiee & M. J. Carey.
An LSM-based Tuple Compaction Framework for Apache AsterixDB - Extended Version
To appear in VLDB, 2020.

[2]

W. Alkowaileet, S. Alsubaiee, M. J. Carey, C. Li, P. Sinthong & W. Wang.
End-to-End Machine Learning with Apache AsterixDB
In Proceedings of the Second Workshop on Data Management for End-To-End Machine Learning, 2018.

[3]

W. Alkowaileet, S. Alsubaiee, M. J. Carey, T. Westmann & Y. Bu.
Large-scale Complex Analytics on Semi-structured Datasets using AsterixDB and Spark
PVLDB, 9(13), 1585-1588, 2016 (Demo)

[4]

W. Alkowaileet, D. Carrillo-Cisneros, D. Lim & I. D. Scherson.
NUMA-aware Multicore Matrix Multiplication
Parallel Processing Letters, 24(04), 1450006, 2014