Hi! I am a Ph.D. candidate advised by Dr. Xun (Steve) Jian in Computer Science department at Virginia Tech. I am a member of the HEAP (High-performance, Energy-efficient, Assured Processing) Lab and I do research in the intersection of Computer Architecture and Operating Systems.
I did my Bachelors from LUMS, Lahore, Pakistan and my Masters from Koç University, Istanbul, Turkey.
Research Interests
My research interests broadly lie in the intersection of Computer Architecture and Operating Systems to perform main memory optimizations.
Experience
Hardware Test Engineer II at Microsoft Azure (January 2021 - June 2021)
Supervisor: Dr. Srilatha Manne and Dr. Pulkit Misra
- Power and performance characterization of machines used in Microsoft Azure to reduce the carbon cost of applications while still achieving reasonable performance.
- Analyzed system and architectural characteristics, such as uncore and core frequencies, and IPC, to determine the impact of each component on total power consumption of a machine.
Summer Research Intern at Microsoft Research (Summer 2020)
Supervisor: Dr. Srilatha Manne and Dr. Esha Choukse
- Studied the sporadic jitter that occurs when multiple VMs with different characteristics run together on the same system.
- Analyzed the benefits of Intel Cache Allocation Technology (CAT) in multi-VM scenarios and argued how Intel CAT does not scale well with processors with high core count running a large number of VMs
Graduate Research Assistant at HEAP Lab at Virginia Tech (Fall 2018 - present)
Supervisor: Dr. Xun Jian
- Instrumenting Linux Kernel to quantize the software overhead of OS-level memory compression.
- Designing and implementing OS features that help to OS-directed hardware memory compression.
Graduate Research Assistant at Koç University (Fall 2016 - Summer 2018)
Supervisor: Dr. Didem Unat
I designed an efficient and intelligent object placement framework for heterogeneous memory systems such as the Intel's Knights Landing (KNL) architecture. The framework enabled applications to reach a speedup of ~2.5x. This work led to 2 publications.
Research Assistant at LUMS (Summer 2015 - Summer 2016)
Supervisor: Dr. Ihsan Ayyub Qazi
I worked on Software-Defined Networks (SDNs) and developed a scheme which allocated network resources efficiently and dynamically in a datacenter.
Publications
- Gagandeep Panwar, Muhammad Laghari, David Bears, Yuqing Liu, Chandler Jearls, Esha Choukse, Kirk W. Cameron, Ali Butt, Xun Jian: Translation-Optimized Memory Compression for Capacity. 55th IEEE/ACM International Symposium on Microarchitecture (MICRO), 2022. [Paper]
- Mohammad Laghari, Najeeb Ahmad, Didem Unat: Phase-Based Data Placement Scheme for Heterogeneous Memory Systems. 30th International Symposium on Computer Architecture and High Performance Computing (SBAC-PAD), 2018. [Paper]
- Mohammad Laghari, Didem Unat: Object Placement for High Bandwidth Memory Augmented with High Capacity Memory. 29th International Symposium on Computer Architecture and High Performance Computing (SBAC-PAD), 2017. [Paper], [Slides]
Others
Here is my CV.
My Masters Thesis titled Data Placement on Heterogeneous Memory Systems can be found here.
Department of Computer Science, Virginia Tech
220 Gilbert St.,
Blacksburg, VA 24060
mlaghari@vt.edu