Bio:
Aditya Kumar is a compiler engineer at Facebook. He has over 7 years of experience in software performance analysis and optimizations, improving the quality of large code bases via tools. His specialties include compiler optimizations, writing novel compiler optimization algorithms, improving efficiency of existing algorithms in widely deployed systems, static analysis, C++ standard libraries. He has implemented multiple novel compiler optimization algorithms in llvm, and gcc. He has also improved several C++ standard library algorithms in libc++, libstdc++; algorithms like string::find and iostream::xsgetn which are some of the most widely used algorithms in the C++ standard library.
His implementation and creation of “Hot cold splitting” in LLVM reduces the startup time of applications on resource-constrained devices, such as the very popular smartphones that are heavily used. Hot cold splitting is also used in the compilation of all iPhone applications, in addition to the entire iOS-operating system. His implementation of GVNHoist in LLVM, an optimization that reduces code size and improves performance of workloads simultaneously, is currently used by major microprocessor vendors like Hewlett Packard and Lenovo to report SPEC benchmark numbers.
He regularly presents at LLVM’s developer conference and serves as program chair of llvm-workshop at CGO since 2017. He has also presented at CppNow-2017, SPLASH-2017 and Design Automation Conference DAC-2017.
He completed his masters from Texas A&M University. His thesis was titled: ‘Rejuvenating C++ programs through demacrofication’. This work has been partially implemented in the well known Visual C++ IDE.
Prior to joining Facebook, he has worked at Qualcomm Innovation Center and Samsung Austin R&D Center as compiler engineer.
Social Media Profiles:
LinkedIn: https://www.linkedin.com/in/hiraditya
Twitter: https://www.twitter.com/_hiraditya_
Available Lectures
To request a single lecture/event, click on the desired lecture and complete the Request Lecture Form.
-
Code Size Compiler Optimizations and Techniques for Embedded Systems
Code size of embedded applications has been a growing concern recently. While storage becomes cheaper and smaller, developers find creative ways to increase code size by adding features or...
- Performance analysis and optimization
With the rise of complex systems that run on a large number of devices on server farms, performance analysis has become ever so important. Having a comprehensive knowledge of different abstraction...- Performance analysis and optimization of C++ standard libraries
C++ standard libraries are some of the most widely running system libraries on consumer devices and server machines. Most engineers assume the performance of standard libraries are as good...To request a tour with this speaker, please complete this online form.
If you are not requesting a tour, click on the desired lecture and complete the Request this Lecture form.
All requests will be sent to ACM headquarters for review.
- Performance analysis and optimization