Single and Multiple Device DSA Problem, Complexities and Online Algorithms
- Computer ScienceISAAC
The goal is to dynamically allocate the job into memory to minimize the usage of space without concurrency and the SRF problem is just a variant of DSA problem.
Upper Bounds for Dynamic Memory Allocation
- Computer ScienceIEEE Transactions on Computers
The upper bound of memory usage for all allocators is derived and a systematic approach to search for allocation/deallocation patterns that might lead to the largest fragmentation is presented.
A constant-time dynamic storage allocator for real-time systems
- Computer ScienceReal-Time Systems
TLSF (Two Level Segregated Fit) is presented, which is designed and implemented to accommodate real-time constraints and exhibits time-bounded behaviour, O(1), and maintains a very good execution time.
Memory management for high-performance applications
- Computer Science
A framework for analyzing and designing high-quality memory managers and a concurrent memory manager called Hoard, which combines the best of both approaches, allowing server applications to manage memory quickly and flexibly while avoiding memory leaks.
Hoard: A Fast, Scalable, and Memory-Efficient Allocator for Shared-MemoryMultiprocessors
- Computer Science
Hoard, a memory allocator for shared-memory multiprocessors, is presented and it is proved that its worst-case memory fragmentation is asymptotically equivalent to that of an optimal uniprocessor allocator.
Study of Energy-Efficient Optimization Techniques for High-Level Homogeneous Resource Management
- Computer ScienceWireless Communications and Mobile Computing
A suite of memory management algorithms that enable dramatic performance improvements at the source code level while allowing for seamless integration across multiple use cases are proposed and presented.
A Fast Data Structure for Dynamic Graphs Based on Hash-Indexed Adjacency Blocks
- Computer ScienceSEA
DHB’s interface is similar to general-purpose abstract graph data types and can be easily used as a drop-in replacement for traditional adjacency arrays and outperforms competing dynamic graph structures for edge insertions, updates, deletions, and traversal operations.
Dynamic memory management in massively parallel systems: a case on GPUs
- Computer ScienceICS
This paper presents a novel approach for designing dynamic memory allocation without a centralized data structure to let threads follow a random search procedure to locate free pages and presents mathematical proofs to demonstrate that the basic random search design achieves asymptotically lower latency than the traditional queue-based design.
Beyond malloc efficiency to fleet efficiency: a hugepage-aware memory allocator
- Computer ScienceOSDI
This work discusses the design and implementation of TEMERAIRE, a hugepage-aware enhancement of TCMALLOC to reduce CPU overheads in the application’s code and presents the results of a 1% experiment at fleet scale as well as the longitudinal rollout in Google's warehouse scale computers.
CoRM: Compactable Remote Memory over RDMA
- Computer ScienceSIGMOD Conference
This work introduces CoRM, an RDMA-accelerated shared memory system that supports memory compaction and ensures strict consistency while providing one-sided RDMA accesses, and shows that CoRM sustains high read throughput during normal operations, comparable to similar systems not providing memory compACTION while experiencing minimal overheads during compaction.