1. X
  2. Ryan Marcus
Log inSign up
Ryan Marcus
679 posts
Image
user avatar
Ryan Marcus
@RyanMarcus
Assistant prof @CIS_Penn. Machine learning for systems, databases. Also discuss.systems/@ryanmarcus
Philadelphia, PA
ryanmarc.us
Joined March 2009
1,091
Following
1,692
Followers
RepliesRepliesMediaMedia
  • user avatar
    Ryan Marcus
    @RyanMarcus
    Dec 26, 2025
    Most database teams optimize what they see in workload logs. But those very optimizations change what users choose to run! In our CIDR paper, we argue that industrial workloads exhibit ๐ฌ๐ฎ๐ซ๐ฏ๐ข๐ฏ๐จ๐ซ๐ฌ๐ก๐ข๐ฉ ๐›๐ข๐š๐ฌ: logs reflect a negotiation between users and the platform.
    A four-step cycle diagram showing feedback between database users and engineers.
โ‘  User submits their workload to the system: A square grid of colored squares represents the workload (3 green, 2 purple).
โ‘ก Engineers observe properties of the workload: A database cylinder leads to a chart showing workload composition: purple 30%, green 60%. A speech bubble from an engineer character says, โ€œMost queries are green. Iโ€™ll trade lower purple performance for higher green performance.โ€
โ‘ข Engineers identify hotspots and optimize their system.
โ‘ฃ Users optimize their workloads based on their platform: A speech bubble from a user character says, โ€œOur platform is good at green, but bad at purple โ€” send more green!โ€ An arrow shows users adjusting the workload and feeding it back into the system.

Overall, the diagram illustrates a feedback loop where system optimizations influence user behavior, which then shapes the workload engineers observe.
  • user avatar
    Ryan Marcus
    @RyanMarcus
    Jun 3, 2025
    OLAP workloads are dominated by repetitive queries -- how can we optimize them? A promising direction is to do ๐—ผ๐—ณ๐—ณ๐—น๐—ถ๐—ป๐—ฒ query optimization, allowing for a much more thorough plan search. Two new SIGMOD papers! ๐Ÿงต
  • user avatar
    Ryan Marcus
    @RyanMarcus
    Feb 15, 2025
    Pair(akeet) programming.
    Image
    00:00
  • user avatar
    Ryan Marcus
    @RyanMarcus
    Jun 7, 2024
    At aiDM@SIGMOD, PhD student Zixuan Yi will present LimeQO, the first *workload-level* learned query optimizer: simultaneously learning to optimize an entire query workload at once! By casting the problem as low rank matrix completion, we show that linear methods are all you need.
    A workload matrix. Each row represents a query, and each column represents a hint. Each entry is the latency of that query using that hint. Some values are unobserved, others are observed.
    The LimeQO system model. A workload matrix is continuously approximated. Then, offline exploration is used to verify which query plans represent improvements.
  • user avatar
    Ryan Marcus
    @RyanMarcus
    May 20, 2024
    Greatly enjoyed talking with Jack! We discussed the "research journey," what it means for DB research to be impactful, and new work from our lab about query optimization!
    user avatar
    Disseminate: The Computer Science Research Podcast
    @DisseminatePod
    May 20, 2024
    ๐Ÿšจ The first episode in our #HighImpact series with Ryan Marcus (@RyanMarcus) is available now! ๐ŸŽง Listen on Spotify โžก๏ธ open.spotify.com/show/6IQIF9oRSโ€ฆ ๐ŸŽง Listen on Apple โžก๏ธ podcasts.apple.com/us/podcast/disโ€ฆ

Log in or sign up for X

See whatโ€™s happening and join the conversation

Continue with phone
or
Log in with username or email
TermsยทPrivacyยทCookiesยทAccessibilityยทAds Infoยทยฉ 2026 X Corp.
Advertisement
Advertisement