Amin Vahdat
Mountain View, California, United States
11K followers
500+ connections
Über uns
At Google, I lead the global Machine Learning, Systems and Cloud AI team, with…
Activity
-
Great CNBC feature on Google and our leadership in custom chips. Tremendous work over many years by Amin Vahdat and powers our leadership in…
Great CNBC feature on Google and our leadership in custom chips. Tremendous work over many years by Amin Vahdat and powers our leadership in…
Liked by Amin Vahdat
-
I have fond memories of learning under Ed Feigenbaum, John McCarthy, Stuart Russell, Doug Lenat and Alan Kay in the 80’s and 90’s, working on the AI…
I have fond memories of learning under Ed Feigenbaum, John McCarthy, Stuart Russell, Doug Lenat and Alan Kay in the 80’s and 90’s, working on the AI…
Liked by Amin Vahdat
-
A nice feature from CNBC on our custom acceleration journey featuring TPUs, VCUs, and Axions. Some great visuals of the labs we work in and the…
A nice feature from CNBC on our custom acceleration journey featuring TPUs, VCUs, and Axions. Some great visuals of the labs we work in and the…
Liked by Amin Vahdat
Erleben Sie
Bildung
Publications
-
Pip: Detecting the Unexpected in Distributed Systems
Proceedings of NSDI
Bugs in distributed systems are often hard to find. Many bugs reflect discrepancies between a system's behavior and the programmer's assumptions about that behavior. We present Pip, an infrastructure for comparing actual behavior and expected behavior to expose structural errors and performance problems in distributed systems. Pip allows programmers to express, in a declarative language, expectations about the system's communications structure, timing, and resource consumption. Pip includes…
Bugs in distributed systems are often hard to find. Many bugs reflect discrepancies between a system's behavior and the programmer's assumptions about that behavior. We present Pip, an infrastructure for comparing actual behavior and expected behavior to expose structural errors and performance problems in distributed systems. Pip allows programmers to express, in a declarative language, expectations about the system's communications structure, timing, and resource consumption. Pip includes system instrumentation and annotation tools to log actual system behavior, and visualization and query tools for exploring expected and unexpected behavior. Pip allows a developer to quickly understand and debug both familiar and unfamiliar systems.
We applied Pip to several applications, including FAB, SplitStream, Bullet, and RanSub. We generated most of the instrumentation for all four applications automatically. We found the needed expectations easy to write, starting in each case with automatically generated expectations. Pip found unexpected behavior in each application, and helped to isolate the causes of poor performance and incorrect behavior.Other authorsSee publication -
WAP5: Black-Box Performance Debugging for Wide-Area Systems
Proceedings of WWW
Wide-area distributed applications are challenging to debug, optimize, and maintain. We present Wide-Area Project 5 (WAP5), which aims to make these tasks easier by exposing the causal structure of communication within an application and by exposing delays that imply bottlenecks. These bottlenecks might not otherwise be obvious, with or without the application's source code. Previous research projects have presented algorithms to reconstruct application structure and the corresponding timing…
Wide-area distributed applications are challenging to debug, optimize, and maintain. We present Wide-Area Project 5 (WAP5), which aims to make these tasks easier by exposing the causal structure of communication within an application and by exposing delays that imply bottlenecks. These bottlenecks might not otherwise be obvious, with or without the application's source code. Previous research projects have presented algorithms to reconstruct application structure and the corresponding timing information from black-box message traces of local-area systems. In this paper we present (1) a new algorithm for reconstructing application structure in both local- and wide-area distributed systems, (2) an infrastructure for gathering application traces in PlanetLab, and (3) our experiences tracing and analyzing three systems: CoDeeN and Coral, two content-distribution networks in PlanetLab; and Slurpee, an enterprise-scale incident-monitoring system.
Other authorsSee publication
More activity by Amin
-
Google makes custom cloud chips, now used to train #Apple AI models as well as its own chatbot Gemini. And it's been making them since 2015 - years…
Google makes custom cloud chips, now used to train #Apple AI models as well as its own chatbot Gemini. And it's been making them since 2015 - years…
Liked by Amin Vahdat
-
Thank you to Katie Tarasov and the team at CNBC for the video and article covering our work on custom silicon. We discuss our AI accelerators…
Thank you to Katie Tarasov and the team at CNBC for the video and article covering our work on custom silicon. We discuss our AI accelerators…
Shared by Amin Vahdat
-
I had the pleasure of collaborating with Partha Ranganathan on a recent blog post discussing how Google designs its own chips to meet the needs of…
I had the pleasure of collaborating with Partha Ranganathan on a recent blog post discussing how Google designs its own chips to meet the needs of…
Liked by Amin Vahdat
-
Institution building now enters the next stage…coming up next, a new School in UC San Diego as a part of the intellectual expansion of a growing…
Institution building now enters the next stage…coming up next, a new School in UC San Diego as a part of the intellectual expansion of a growing…
Liked by Amin Vahdat
-
Yanbing Li's expertise in AI, machine learning, and cloud infrastructure aligns seamlessly with Datadog's mission to deliver unified, real-time…
Yanbing Li's expertise in AI, machine learning, and cloud infrastructure aligns seamlessly with Datadog's mission to deliver unified, real-time…
Liked by Amin Vahdat
-
It's been truly touching to see all of the tributes remembering Susan Wojcicki. She was a once in a lifetime leader who changed my life, like she did…
It's been truly touching to see all of the tributes remembering Susan Wojcicki. She was a once in a lifetime leader who changed my life, like she did…
Liked by Amin Vahdat
-
I am still processing the news of Susan Wojcicki's passing. I had the privilege of meeting Susan through women's events at Google and in meetings…
I am still processing the news of Susan Wojcicki's passing. I had the privilege of meeting Susan through women's events at Google and in meetings…
Liked by Amin Vahdat
-
Increasing access to quality education is close to my heart, and something that has been a big focus throughout my career - including my time at…
Increasing access to quality education is close to my heart, and something that has been a big focus throughout my career - including my time at…
Liked by Amin Vahdat
-
Susan was the best and most empathetic leader I've been around. I was excited to move to Google (Cloud followed by YouTube) because of her and doubly…
Susan was the best and most empathetic leader I've been around. I was excited to move to Google (Cloud followed by YouTube) because of her and doubly…
Liked by Amin Vahdat
-
Susan Wojcicki’s passing is an unimaginable tragedy. I am still numb and trying to process it. It just seems wrong for such a truly GOOD person to be…
Susan Wojcicki’s passing is an unimaginable tragedy. I am still numb and trying to process it. It just seems wrong for such a truly GOOD person to be…
Liked by Amin Vahdat
-
Like countless others, I'm heartbroken that Susan Wojcicki has passed - for her, for her family, for all those who she inspired, and for the…
Like countless others, I'm heartbroken that Susan Wojcicki has passed - for her, for her family, for all those who she inspired, and for the…
Liked by Amin Vahdat
Other similar profiles
Gemeinsame Artikel erkunden
We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.
Explore More