2000
DOI: 10.1145/342001.339666
|View full text |Cite
|
Sign up to set email alerts
|

Recency-based TLB preloading

Abstract: Caching and other latency tolerating techniques have been quite successful in maintaining high memory system performance for general purpose processors. However, TLB misses have become a serious bottleneck as working sets are growing beyond the capacity of TLBs. This work presents one of the first attempts to hide TLB miss latency by using preloading techniques. We present results for traditional next-page TLB miss preloading - an approach shown to cut some of the misses. Howe… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
60
0

Year Published

2005
2005
2020
2020

Publication Types

Select...
4
3
2

Relationship

0
9

Authors

Journals

citations
Cited by 32 publications
(60 citation statements)
references
References 16 publications
0
60
0
Order By: Relevance
“…In order to reduce TLB misses, the following methods have been proposed: utilizing TLB entries efficiently by enlarging the page size [9], pre-registering the information in the page table by estimating TLB misses [10], and scheduling processes to prevent TLB misses [11]. In [9], TLB misses are reduced by reducing the number of TLB entries per memory size, whereas in [10] registers the memory use in the TLB entry is registered before the memory is accessed.…”
Section: Related Workmentioning
confidence: 99%
See 1 more Smart Citation
“…In order to reduce TLB misses, the following methods have been proposed: utilizing TLB entries efficiently by enlarging the page size [9], pre-registering the information in the page table by estimating TLB misses [10], and scheduling processes to prevent TLB misses [11]. In [9], TLB misses are reduced by reducing the number of TLB entries per memory size, whereas in [10] registers the memory use in the TLB entry is registered before the memory is accessed.…”
Section: Related Workmentioning
confidence: 99%
“…In [9], TLB misses are reduced by reducing the number of TLB entries per memory size, whereas in [10] registers the memory use in the TLB entry is registered before the memory is accessed. In contrast, the proposed method manages a specific area by using TLB entries without using page tables.…”
Section: Related Workmentioning
confidence: 99%
“…Saulsbury et al [23] proposed a recency-based TLB preloading. It maintains the TLB information in a Mattson stack, and preloads adjacent entries in the stack upon a TLB miss.…”
Section: Related Workmentioning
confidence: 99%
“…Additionally, several solutions have been proposed to minimize TLB miss rate and latency, and thus improve TLB performance: tunning TLB size or associativity [20], multiple TLB levels, different page sizes (superpages) [82], or prefetching [75]. However, most of these proposal targeted uniprocessors.…”
Section: Boosting Tlb Performancementioning
confidence: 99%