Memory management and address translation need significant optimizations in order to not behindrances in the near future. Currently, plenty of work has started to address issues within the current abstraction of the hardware-software codesign of paging. I argue that a new abstraction is needed in order to properly address this growing issue. I conducted an investigation into reviving software-based abstractions to manage memory with through creating a Tool to Examine and X’form Allocation State, or TEXAS. TEXAS is a software-software codesign to manage memory using the compiler and runtime. There are several projects that explore TEXAS’s capabilities for performing software-basedmemory management, such as: CARAT CAKE, which uses TEXAS to remove the need for virtual memory along with the supporting hardware like the TLB; CARMOT, which uses TEXAS to provide recommendations to programmers attempting to use complex language extensions like OpenMP; And finally Manufacturing Locality, which uses TEXAS to dynamically convert the temporal access patterns of programs into spatial locality. These projects, and affiliated projects, show the potential of software-based memory management’s bright future.

Creator

• Suchy, Brian

DOI

• https://doi.org/10.21985/n2-9kn8-a434

Subject

• Computer engineering
• Computer science

Language

• en

Alternate Identifier

• http://dissertations.umi.com/northwestern:16132
• etdadmin_upload_912710

Keyword

• Runtime Environments
• Operating Systems
• Microarchitecture
• Blended Systems
• Compilers

Date created

• 2022-01-01

Resource type

• Dissertation

Rights statement

• In Copyright