EDC: An Elastic Data Cache to Optimizing the I/O Performance in Deduplicated SSDs

Author(s):  
Mengting Lu ◽  
Fang Wang ◽  
Zongwei Li ◽  
Wenpeng He
Keyword(s):  
2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Costanza Bonadiman ◽  
Valentina Brombin ◽  
Giovanni B. Andreozzi ◽  
Piera Benna ◽  
Massimo Coltorti ◽  
...  

AbstractThe occurrence of phlogopite and amphibole in mantle ultramafic rocks is widely accepted as the modal effect of metasomatism in the upper mantle. However, their simultaneous formation during metasomatic events and the related sub-solidus equilibrium with the peridotite has not been extensively studied. In this work, we discuss the geochemical conditions at which the pargasite-phlogopite assemblage becomes stable, through the investigation of two mantle xenoliths from Mount Leura (Victoria State, Australia) that bear phlogopite and the phlogopite + amphibole (pargasite) pair disseminated in a harzburgite matrix. Combining a mineralogical study and thermodynamic modelling, we predict that the P–T locus of the equilibrium reaction pargasite + forsterite = Na-phlogopite + 2 diopside + spinel, over the range 1.3–3.0 GPa/540–1500 K, yields a negative Clapeyron slope of -0.003 GPa K–1 (on average). The intersection of the P–T locus of supposed equilibrium with the new mantle geotherm calculated in this work allowed us to state that the Mount Leura xenoliths achieved equilibrium at 2.3 GPa /1190 K, that represents a plausible depth of ~ 70 km. Metasomatic K-Na-OH rich fluids stabilize hydrous phases. This has been modelled by the following equilibrium equation: 2 (K,Na)-phlogopite + forsterite = 7/2 enstatite + spinel + fluid (components: Na2O,K2O,H2O). Using quantum-mechanics, semi-empirical potentials, lattice dynamics and observed thermo-elastic data, we concluded that K-Na-OH rich fluids are not effective metasomatic agents to convey alkali species across the upper mantle, as the fluids are highly reactive with the ultramafic system and favour the rapid formation of phlogopite and amphibole. In addition, oxygen fugacity estimates of the Mount Leura mantle xenoliths [Δ(FMQ) = –1.97 ± 0.35; –1.83 ± 0.36] indicate a more reducing mantle environment than what is expected from the occurrence of phlogopite and amphibole in spinel-bearing peridotites. This is accounted for by our model of full molecular dissociation of the fluid and incorporation of the O-H-K-Na species into (OH)-K-Na-bearing mineral phases (phlogopite and amphibole), that leads to a peridotite metasomatized ambient characterized by reduced oxygen fugacity.


2021 ◽  
Vol 18 (3) ◽  
pp. 1-22
Author(s):  
Michael Stokes ◽  
David Whalley ◽  
Soner Onder

While data filter caches (DFCs) have been shown to be effective at reducing data access energy, they have not been adopted in processors due to the associated performance penalty caused by high DFC miss rates. In this article, we present a design that both decreases the DFC miss rate and completely eliminates the DFC performance penalty even for a level-one data cache (L1 DC) with a single cycle access time. First, we show that a DFC that lazily fills each word in a DFC line from an L1 DC only when the word is referenced is more energy-efficient than eagerly filling the entire DFC line. For a 512B DFC, we are able to eliminate loads of words into the DFC that are never referenced before being evicted, which occurred for about 75% of the words in 32B lines. Second, we demonstrate that a lazily word filled DFC line can effectively share and pack data words from multiple L1 DC lines to lower the DFC miss rate. For a 512B DFC, we completely avoid accessing the L1 DC for loads about 23% of the time and avoid a fully associative L1 DC access for loads 50% of the time, where the DFC only requires about 2.5% of the size of the L1 DC. Finally, we present a method that completely eliminates the DFC performance penalty by speculatively performing DFC tag checks early and only accessing DFC data when a hit is guaranteed. For a 512B DFC, we improve data access energy usage for the DTLB and L1 DC by 33% with no performance degradation.


Author(s):  
B. Shameedha Begum ◽  
N. Ramasubramanian

Embedded systems are designed for a variety of applications ranging from Hard Real Time applications to mobile computing, which demands various types of cache designs for better performance. Since real-time applications place stringent requirements on performance, the role of the cache subsystem assumes significance. Reconfigurable caches meet performance requirements under this context. Existing reconfigurable caches tend to use associativity and size for maximizing cache performance. This article proposes a novel approach of a reconfigurable and intelligent data cache (L1) based on replacement algorithms. An intelligent embedded data cache and a dynamic reconfigurable intelligent embedded data cache have been implemented using Verilog 2001 and tested for cache performance. Data collected by enabling the cache with two different replacement strategies have shown that the hit rate improves by 40% when compared to LRU and 21% when compared to MRU for sequential applications which will significantly improve performance of embedded real time application.


2008 ◽  
Vol 5 (19) ◽  
pp. 833-839
Author(s):  
Seungmin Jung ◽  
Hyotaek Shim ◽  
Seungryoul Maeng

Integration ◽  
2012 ◽  
Vol 45 (3) ◽  
pp. 237-245 ◽  
Author(s):  
Azam Seyedi ◽  
Adrià Armejach ◽  
Adrián Cristal ◽  
Osman S. Unsal ◽  
Ibrahim Hur ◽  
...  
Keyword(s):  

Sign in / Sign up

Export Citation Format

Share Document