Bibliografie tematiche / Disaggregated Memory

Letteratura scientifica selezionata sul tema "Disaggregated Memory"

Autore: Grafiati
Pubblicato: 1 giugno 2024

Cita una fonte nei formati APA, MLA, Chicago, Harvard e in molti altri stili

Scegli il tipo di fonte:

Consulta la lista di attuali articoli, libri, tesi, atti di convegni e altre fonti scientifiche attinenti al tema "Disaggregated Memory".

Accanto a ogni fonte nell'elenco di riferimenti c'è un pulsante "Aggiungi alla bibliografia". Premilo e genereremo automaticamente la citazione bibliografica dell'opera scelta nello stile citazionale di cui hai bisogno: APA, MLA, Harvard, Chicago, Vancouver ecc.

Puoi anche scaricare il testo completo della pubblicazione scientifica nel formato .pdf e leggere online l'abstract (il sommario) dell'opera se è presente nei metadati.

Indice

  1. Articoli di riviste
  2. Tesi
  3. Capitoli di libri
  4. Atti di convegni
  5. Rapporti di organizzazioni

Articoli di riviste sul tema "Disaggregated Memory":

1

Cao, Wenqi, e Ling Liu. "Hierarchical Orchestration of Disaggregated Memory". IEEE Transactions on Computers 69, n. 6 (1 giugno 2020): 844–55. http://dx.doi.org/10.1109/tc.2020.2968525.

Testo completo
Gli stili APA, Harvard, Vancouver, ISO e altri
2

Calciu, Irina, M. Talha Imran, Ivan Puddu, Sanidhya Kashyap, Hasan Al Maruf, Onur Mutlu e Aasheesh Kolli. "Using Local Cache Coherence for Disaggregated Memory Systems". ACM SIGOPS Operating Systems Review 57, n. 1 (26 giugno 2023): 21–28. http://dx.doi.org/10.1145/3606557.3606561.

Testo completo
Gli stili APA, Harvard, Vancouver, ISO e altri
Abstract (sommario):
Disaggregated memory provides many cost savings and resource provisioning benefits for current datacenters, but software systems enabling disaggregated memory access result in high performance penalties. These systems require intrusive code changes to port applications for disaggregated memory or employ slow virtual memory mechanisms to avoid code changes. Such mechanisms result in high overhead page faults to access remote data and high dirty data amplification when tracking changes to cached data at page-granularity. In this paper, we propose a fundamentally new approach for disaggregated memory systems, based on the observation that we can use local cache coherence to track applications' memory accesses transparently, without code changes, at cache-line granularity. This simple idea (1) eliminates page faults from the application critical path when accessing remote data, and (2) decouples the application memory access tracking from the virtual memory page size, enabling cache-line granularity dirty data tracking and eviction. Using this observation, we implemented a new software runtime for disaggregated memory that improves average memory access time and reduces dirty data amplification1.
3

Maruf, Hasan Al, Yuhong Zhong, Hongyi Wang, Mosharaf Chowdhury, Asaf Cidon e Carl Waldspurger. "Memtrade: Marketplace for Disaggregated Memory Clouds". ACM SIGMETRICS Performance Evaluation Review 51, n. 1 (26 giugno 2023): 1–2. http://dx.doi.org/10.1145/3606376.3593553.

Testo completo
Gli stili APA, Harvard, Vancouver, ISO e altri
Abstract (sommario):
We present Memtrade, the first practical marketplace for disaggregated memory clouds. Clouds introduce a set of unique challenges for resource disaggregation across different tenants, including resource harvesting, isolation, and matching. Memtrade allows producer virtual machines (VMs) to lease both their unallocated memory and allocated-but-idle application memory to remote consumer VMs for a limited period of time. Memtrade does not require any modifications to host-level system software or support from the cloud provider. It harvests producer memory using an application-aware control loop to form a distributed transient remote memory pool with minimal performance impact; it employs a broker to match producers with consumers while satisfying performance constraints; and it exposes the matched memory to consumers through different abstractions. As a proof of concept, we propose two such memory access interfaces for Memtrade consumers -- a transient KV cache for specified applications and a swap interface that is application-transparent. Our evaluation shows that Memtrade provides significant performance benefits for consumers (improving average read latency up to 2.8X) while preserving confidentiality and integrity, with little impact on producer applications (degrading performance by less than 2.1%).
4

Maruf, Hasan Al, Yuhong Zhong, Hongyi Wang, Mosharaf Chowdhury, Asaf Cidon e Carl Waldspurger. "Memtrade: Marketplace for Disaggregated Memory Clouds". Proceedings of the ACM on Measurement and Analysis of Computing Systems 7, n. 2 (19 maggio 2023): 1–27. http://dx.doi.org/10.1145/3589985.

Testo completo
Gli stili APA, Harvard, Vancouver, ISO e altri
Abstract (sommario):
We present Memtrade, the first practical marketplace for disaggregated memory clouds. Clouds introduce a set of unique challenges for resource disaggregation across different tenants, including resource harvesting, isolation, and matching. Memtrade allows producer virtual machines (VMs) to lease both their unallocated memory and allocated-but-idle application memory to remote consumer VMs for a limited period of time. Memtrade does not require any modifications to host-level system software or support from the cloud provider. It harvests producer memory using an application-aware control loop to form a distributed transient remote memory pool with minimal performance impact; it employs a broker to match producers with consumers while satisfying performance constraints; and it exposes the matched memory to consumers through different abstractions. As a proof of concept, we propose two such memory access interfaces for Memtrade consumers -- a transient KV cache for specified applications and a swap interface that is application-transparent. Our evaluation using real-world cluster traces shows that Memtrade provides significant performance benefit for consumers (improving average read latency up to 2.8X) while preserving confidentiality and integrity, with little impact on producer applications (degrading performance by less than 2.1%).
5

Min, Xinhao, Kai Lu, Pengyu Liu, Jiguang Wan, Changsheng Xie, Daohui Wang, Ting Yao e Huatao Wu. "SepHash: A Write-Optimized Hash Index On Disaggregated Memory via Separate Segment Structure". Proceedings of the VLDB Endowment 17, n. 5 (gennaio 2024): 1091–104. http://dx.doi.org/10.14778/3641204.3641218.

Testo completo
Gli stili APA, Harvard, Vancouver, ISO e altri
Abstract (sommario):
Disaggregated memory separates compute and memory resources into independent pools connected by fast RDMA (Remote Direct Memory Access) networks, which can improve memory utilization, reduce cost, and enable elastic scaling of compute and memory resources. Hash indexes provide high-performance single-point operations and are widely used in distributed systems and databases. However, under disaggregated memory, existing hash indexes suffer from write performance degradation due to high resize overhead and concurrency control overhead. Traditional write-optimized hash indexes are not efficient for disaggregated memory and sacrifice read performance. In this paper, we propose SepHash, a write-optimized hash index for disaggregated memory. First, SepHash proposes a two-level separate segment structure that significantly reduces the bandwidth consumption of resize operations. Second, SepHash employs a low-latency concurrency control strategy to eliminate unnecessary mutual exclusion and check overhead during insert operations. Finally, SepHash designs an efficient cache and filter to accelerate read operations. The evaluation results show that, compared to state-of-the-art distributed hash indexes, SepHash achieves a 3.3X higher write performance while maintaining comparable read performance.
6

Ishizaki, Teruaki, e Yoshiro Yamabe. "Memory-centric Architecture for Disaggregated Computers". NTT Technical Review 19, n. 7 (luglio 2021): 65–69. http://dx.doi.org/10.53829/ntr202107fa9.

Testo completo
Gli stili APA, Harvard, Vancouver, ISO e altri
7

Koo, Bonmoo, Jaesang Hwang, Jonghyeok Park e Wook-Hee Kim. "Converting Concurrent Range Index Structure to Range Index Structure for Disaggregated Memory". Applied Sciences 13, n. 20 (10 ottobre 2023): 11130. http://dx.doi.org/10.3390/app132011130.

Testo completo
Gli stili APA, Harvard, Vancouver, ISO e altri
Abstract (sommario):
In this work, we propose the Spread approach, which tailors a concurrent range index structure to a range index structure for disaggregated memory connected via RDMA (Remote Direct Memory Access). The Spread approach leverages the concept of tolerating transient inconsistencies in a concurrent range index structure to reduce the amount of expensive RDMA operations. Based on the Spread approach, we converted Blink-tree, a concurrent range index structure, to a range index structure for disaggregated memory called RF-tree. In our experimental study, RF-tree shows comparable performance to Sherman, a state-of-the-art and carefully crafted range index structure for disaggregated memory.
8

Alachiotis, Nikolaos, Panagiotis Skrimponis, Manolis Pissadakis e Dionisios Pnevmatikatos. "Scalable Phylogeny Reconstruction with Disaggregated Near-memory Processing". ACM Transactions on Reconfigurable Technology and Systems 15, n. 3 (30 settembre 2022): 1–32. http://dx.doi.org/10.1145/3484983.

Testo completo
Gli stili APA, Harvard, Vancouver, ISO e altri
Abstract (sommario):
Disaggregated computer architectures eliminate resource fragmentation in next-generation datacenters by enabling virtual machines to employ resources such as CPUs, memory, and accelerators that are physically located on different servers. While this paves the way for highly compute- and/or memory-intensive applications to potentially deploy all CPUs and/or memory resources in a datacenter, it poses a major challenge to the efficient deployment of hardware accelerators: input/output data can reside on different servers than the ones hosting accelerator resources, thereby requiring time- and energy-consuming remote data transfers that diminish the gains of hardware acceleration. Targeting a disaggregated datacenter architecture similar to the IBM dReDBox disaggregated datacenter prototype, the present work explores the potential of deploying custom acceleration units adjacently to the disaggregated-memory controller on memory bricks (in dReDBox terminology), which is implemented on FPGA technology, to reduce data movement and improve performance and energy efficiency when reconstructing large phylogenies (evolutionary relationships among organisms). A fundamental computational kernel is the Phylogenetic Likelihood Function (PLF), which dominates the total execution time (up to 95%) of widely used maximum-likelihood methods. Numerous efforts to boost PLF performance over the years focused on accelerating computation; since the PLF is a data-intensive, memory-bound operation, performance remains limited by data movement, and memory disaggregation only exacerbates the problem. We describe two near-memory processing models, one that addresses the problem of workload distribution to memory bricks, which is particularly tailored toward larger genomes (e.g., plants and mammals), and one that reduces overall memory requirements through memory-side data interpolation transparently to the application, thereby allowing the phylogeny size to scale to a larger number of organisms without requiring additional memory.
9

Jeong, Yeonwoo, Gyeonghwan Jung, Kyuli Park, Youngjae Kim e Sungyong Park. "Empirical Analysis of Disaggregated Cloud Memory on Memory Intensive Applications". JOURNAL OF SEMICONDUCTOR TECHNOLOGY AND SCIENCE 23, n. 5 (31 ottobre 2023): 273–82. http://dx.doi.org/10.5573/jsts.2023.23.5.273.

Testo completo
Gli stili APA, Harvard, Vancouver, ISO e altri
10

Gonzalez, Jorge, Mauricio G. Palma, Maarten Hattink, Ruth Rubio-Noriega, Lois Orosa, Onur Mutlu, Keren Bergman e Rodolfo Azevedo. "Optically connected memory for disaggregated data centers". Journal of Parallel and Distributed Computing 163 (maggio 2022): 300–312. http://dx.doi.org/10.1016/j.jpdc.2022.01.013.

Testo completo
Gli stili APA, Harvard, Vancouver, ISO e altri
Più fonti
Ti possono interessare anche gli elenchi estesi di opere sul tema "Disaggregated Memory":
Articoli di riviste

Tesi sul tema "Disaggregated Memory":

1

Bielski, Maciej. "Nouvelles techniques de virtualisation de la mémoire et des entrées-sorties vers les périphériques pour les prochaines générations de centres de traitement de données basés sur des équipements répartis déstructurés". Thesis, Université Paris-Saclay (ComUE), 2019. http://www.theses.fr/2019SACLT022/document.

Testo completo
Gli stili APA, Harvard, Vancouver, ISO e altri
Abstract (sommario):
Cette thèse s'inscrit dans le contexte de la désagrégation des systèmes informatiques - une approche novatrice qui devrait gagner en popularité dans le secteur des centres de données. A la différence des systèmes traditionnels en grappes, où les ressources sont fournies par une ou plusieurs machines, dans les systèmes désagrégés les ressources sont fournies par des nœuds discrets, chaque nœud ne fournissant qu'un seul type de ressources (unités centrales de calcul, mémoire, périphériques). Au lieu du terme de machine, le terme de créneau (slot) est utilisé pour décrire une unité de déploiement de charge de travail. L'emplacement est assemblé dynamiquement avant un déploiement de charge de travail par l'orchestrateur système.Dans l'introduction nous abordons le sujet de la désagrégation et en présentons les avantages par rapport aux architectures en grappes. Nous ajoutons également au tableau une couche de virtualisation car il s'agit d'un élément crucial des centres de données. La virtualisation fournit une isolation entre les charges de travail déployées et un partitionnement flexible des ressources. Elle doit cependant être adaptée afin de tirer pleinement parti de la désagrégation. C'est pourquoi les principales contributions de ce travail se concentrent sur la prise en charge de la couche de virtualisation pour la mémoire désagrégée et la mise à disposition des périphériques.La première contribution principale présente les modifications de la pile logicielle liées au redimensionnement flexible de la mémoire d'une machine virtuelle (VM). Elles permettent d'ajuster la quantité de RAM hébergée (c'est à dire utilisée par la charge de travail en cours d'exécution dans une VM) pendant l'exécution avec une granularité d'une section mémoire. Du point de vue du logiciel il est transparent que la RAM proviennent de banques de mémoire locales ou distantes.La deuxième contribution discute des notions de partage de mémoire entre machines virtuelles et de migration des machines virtuelles dans le contexte de la désagrégation. Nous présentons d'abord comment des régions de mémoire désagrégées peuvent être partagées entre des machines virtuelles fonctionnant sur différents nœuds. De plus, nous discutons des différentes variantes de la méthode de sérialisation des accès simultanés. Nous expliquons ensuite que la notion de migration de VM a acquis une double signification avec la désagrégation. En raison de la désagrégation des ressources, une charge de travail est associée au minimum à un nœud de calcul et a un nœud mémoire. Il est donc possible qu'elle puisse être migrée vers des nœuds de calcul différents tout en continuant à utiliser la même mémoire, ou l'inverse. Nous discutons des deux cas et décrivons comment cela peut ouvrir de nouvelles opportunités pour la consolidation des serveurs.La dernière contribution de cette thèse est liée à la virtualisation des périphériques désagrégés. Partant de l'hypothèse que la désagrégation de l'architecture apporte de nombreux effets positifs en général, nous expliquons pourquoi elle n'est pas immédiatement compatible avec la technique d'attachement direct, est pourtant très populaire pour sa performance quasi native. Pour remédier à cette limitation, nous présentons une solution qui adapte le concept d'attachement direct à la désagrégation de l'architecture. Grâce à cette solution, les dispositifs désagrégés peuvent être directement attachés aux machines virtuelles, comme s'ils étaient branchés localement. De plus, l'OS hébergé, pour lequel la configuration de l'infrastructure sous-jacente n'est pas visible, n'est pas lui-même concerné par les modifications introduites
This dissertation is positioned in the context of the system disaggregation - a novel approach expected to gain popularity in the data center sector. In traditional clustered systems resources are provided by one or multiple machines. Differently to that, in disaggregated systems resources are provided by discrete nodes, each node providing only one type of resources (CPUs, memory and peripherals). Instead of a machine, the term of a slot is used to describe a workload deployment unit. The slot is dynamically assembled before a workload deployment by the unit called system orchestrator.In the introduction of this work, we discuss the subject of disaggregation and present its benefits, compared to clustered architectures. We also add a virtualization layer to the picture as it is a crucial part of data center systems. It provides an isolation between deployed workloads and a flexible resources partitioning. However, the virtualization layer needs to be adapted in order to take full advantage of disaggregation. Thus, the main contributions of this work are focused on the virtualization layer support for disaggregated memory and devices provisioning.The first main contribution presents the software stack modifications related to flexible resizing of a virtual machine (VM) memory. They allow to adjust the amount of guest (running in a VM) RAM at runtime on a memory section granularity. From the software perspective it is transparent whether they come from local or remote memory banks.As a second main contribution we discuss the notions of inter-VM memory sharing and VM migration in the disaggregation context. We first present how regions of disaggregated memory can be shared between VMs running on different nodes. This sharing is performed in a way that involved guests which are not aware of the fact that they are co-located on the same computing node or not. Additionally, we discuss different flavors of concurrent accesses serialization methods. We then explain how the VM migration term gained a twofold meaning. Because of resources disaggregation, a workload is associated to at least one computing node and one memory node. It is therefore possible that it is migrated to a different computing node and keeps using the same memory, or the opposite. We discuss both cases and describe how this can open new opportunities for server consolidation.The last main contribution of this dissertation is related to disaggregated peripherals virtualization. Starting from the assumption that the architecture disaggregation brings many positive effects in general, we explain why it breaks the passthrough peripheral attachment technique (also known as a direct attachment), which is very popular for its near-native performance. To address this limitation we present a design that adapts the passthrough attachment concept to the architecture disaggregation. By this novel design, disaggregated devices can be directly attached to VMs, as if they were plugged locally. Moreover, all modifications do not involve the guest OS itself, for which the setup of the underlying infrastructure is not visible
2

Pipereau, Yohan. "Improving memory usage in virtual machines". Electronic Thesis or Diss., Institut polytechnique de Paris, 2023. http://www.theses.fr/2023IPPAS021.

Testo completo
Gli stili APA, Harvard, Vancouver, ISO e altri
Abstract (sommario):
Les centres de données reposent largement sur les machines virtuelles comme unité de déploiement afin de garantir une isolation forte entre deux déploiements. L'introduction de la virtualisation a permis une amélioration significative de l'utilisation des ressources d'un datacenter par rapport à la location d'une machine physique pour chaque utilisateur. Malgré ce gain, les machines virtuelles offrent toujours un gain d'utilisation de ressources plus faible que l'isolation fournie par le système d'exploitation à travers des processus. Deux phénomènes expliquent le gain relatif des machines virtuelles : au moment de l'allocation d'une VM, un orchestrateur traduit l'allocation en une requête d'allocation de plusieurs ressources sur plusieurs serveurs. Le problème que doit résoudre l'orchestrateur afin de trouver un placement optimal des VMs est NP-difficile, imposant à l'orchestrateur l'utilisation d'heuristiques de placement qui laissent une part de ressource inutilisées sur chaque serveur. De plus, durant l'exécution de la VM, la consommation mémoire augmente à mesure que les accès sont effectués et la différence entre allocation mémoire et utilisation mémoire résulte en un ensemble de ressource mémoire qui ne peuvent être utilisé sans risque de crash de VMs. Dans un premier temps, nous proposons un prototype permettant d'accéder aux ressources mémoires inutilisées sur d'autres serveurs. Notre solution est transparente pour les applications s'exécutant dans une VM et offre une réservation à grain fin des ressources mémoires distantes. Dans un second temps, nous présentons les résultats de notre étude des techniques existantes permettant d'ajuster la capacité mémoire d'une VM sur la mémoire qu'elle utilise. Notre travail montre que les solutions actuelles implémentées au niveau de l'hyperviseur introduisent une dégradation des performances des VMs ainsi que des temps de réponse élevés empêchant le partage des ressources mémoires par plusieurs VMs. Nous proposons une solution utilisant les informations disponibles dans la VM pour adapter rapidement leurs capacités mémoires
Data-centers rely on virtual machines (VMs) to offer isolation between deployments.While, the use of VMs enables better resource usage compared to running a service per bare-metal machine, it achieves poorer resource usage than multi-processes solutions.This is caused by two phenomenon:At VM allocation time, VMs are scheduled as resource requests on a VM scheduler which perform virtual machine allocations across a set of servers.Optimal solution to this scheduling problem is NP-hard leading to the adoption of heuristic based allocation that let a good percentage of unallocated memory on each servers known as `stranded memory`.At VM runtime, VM memory is consumed on-demand and the difference between memory allocation and usage results in a decent portion of `allocated unused memory` currently impractically usable.First, we propose a transparent solution for applications running inside VMs to remotely access stranded memory in remote machines with fine-grained reservation of remote resources.Second, we review current techniques trying to fit allocated memory to used memory.We show that all these techniques are managed by the hypervisor and introduce performance degradation in VMs and more importantly high response time which makes resource sharing unpractical.Instead, we propose an abstraction to perform VM-initiated memory provisioning and we present early result of fast adaptation of VM memory
3

Bielski, Maciej. "Nouvelles techniques de virtualisation de la mémoire et des entrées-sorties vers les périphériques pour les prochaines générations de centres de traitement de données basés sur des équipements répartis déstructurés". Electronic Thesis or Diss., Université Paris-Saclay (ComUE), 2019. http://www.theses.fr/2019SACLT022.

Testo completo
Gli stili APA, Harvard, Vancouver, ISO e altri
Abstract (sommario):
Cette thèse s'inscrit dans le contexte de la désagrégation des systèmes informatiques - une approche novatrice qui devrait gagner en popularité dans le secteur des centres de données. A la différence des systèmes traditionnels en grappes, où les ressources sont fournies par une ou plusieurs machines, dans les systèmes désagrégés les ressources sont fournies par des nœuds discrets, chaque nœud ne fournissant qu'un seul type de ressources (unités centrales de calcul, mémoire, périphériques). Au lieu du terme de machine, le terme de créneau (slot) est utilisé pour décrire une unité de déploiement de charge de travail. L'emplacement est assemblé dynamiquement avant un déploiement de charge de travail par l'orchestrateur système.Dans l'introduction nous abordons le sujet de la désagrégation et en présentons les avantages par rapport aux architectures en grappes. Nous ajoutons également au tableau une couche de virtualisation car il s'agit d'un élément crucial des centres de données. La virtualisation fournit une isolation entre les charges de travail déployées et un partitionnement flexible des ressources. Elle doit cependant être adaptée afin de tirer pleinement parti de la désagrégation. C'est pourquoi les principales contributions de ce travail se concentrent sur la prise en charge de la couche de virtualisation pour la mémoire désagrégée et la mise à disposition des périphériques.La première contribution principale présente les modifications de la pile logicielle liées au redimensionnement flexible de la mémoire d'une machine virtuelle (VM). Elles permettent d'ajuster la quantité de RAM hébergée (c'est à dire utilisée par la charge de travail en cours d'exécution dans une VM) pendant l'exécution avec une granularité d'une section mémoire. Du point de vue du logiciel il est transparent que la RAM proviennent de banques de mémoire locales ou distantes.La deuxième contribution discute des notions de partage de mémoire entre machines virtuelles et de migration des machines virtuelles dans le contexte de la désagrégation. Nous présentons d'abord comment des régions de mémoire désagrégées peuvent être partagées entre des machines virtuelles fonctionnant sur différents nœuds. De plus, nous discutons des différentes variantes de la méthode de sérialisation des accès simultanés. Nous expliquons ensuite que la notion de migration de VM a acquis une double signification avec la désagrégation. En raison de la désagrégation des ressources, une charge de travail est associée au minimum à un nœud de calcul et a un nœud mémoire. Il est donc possible qu'elle puisse être migrée vers des nœuds de calcul différents tout en continuant à utiliser la même mémoire, ou l'inverse. Nous discutons des deux cas et décrivons comment cela peut ouvrir de nouvelles opportunités pour la consolidation des serveurs.La dernière contribution de cette thèse est liée à la virtualisation des périphériques désagrégés. Partant de l'hypothèse que la désagrégation de l'architecture apporte de nombreux effets positifs en général, nous expliquons pourquoi elle n'est pas immédiatement compatible avec la technique d'attachement direct, est pourtant très populaire pour sa performance quasi native. Pour remédier à cette limitation, nous présentons une solution qui adapte le concept d'attachement direct à la désagrégation de l'architecture. Grâce à cette solution, les dispositifs désagrégés peuvent être directement attachés aux machines virtuelles, comme s'ils étaient branchés localement. De plus, l'OS hébergé, pour lequel la configuration de l'infrastructure sous-jacente n'est pas visible, n'est pas lui-même concerné par les modifications introduites
This dissertation is positioned in the context of the system disaggregation - a novel approach expected to gain popularity in the data center sector. In traditional clustered systems resources are provided by one or multiple machines. Differently to that, in disaggregated systems resources are provided by discrete nodes, each node providing only one type of resources (CPUs, memory and peripherals). Instead of a machine, the term of a slot is used to describe a workload deployment unit. The slot is dynamically assembled before a workload deployment by the unit called system orchestrator.In the introduction of this work, we discuss the subject of disaggregation and present its benefits, compared to clustered architectures. We also add a virtualization layer to the picture as it is a crucial part of data center systems. It provides an isolation between deployed workloads and a flexible resources partitioning. However, the virtualization layer needs to be adapted in order to take full advantage of disaggregation. Thus, the main contributions of this work are focused on the virtualization layer support for disaggregated memory and devices provisioning.The first main contribution presents the software stack modifications related to flexible resizing of a virtual machine (VM) memory. They allow to adjust the amount of guest (running in a VM) RAM at runtime on a memory section granularity. From the software perspective it is transparent whether they come from local or remote memory banks.As a second main contribution we discuss the notions of inter-VM memory sharing and VM migration in the disaggregation context. We first present how regions of disaggregated memory can be shared between VMs running on different nodes. This sharing is performed in a way that involved guests which are not aware of the fact that they are co-located on the same computing node or not. Additionally, we discuss different flavors of concurrent accesses serialization methods. We then explain how the VM migration term gained a twofold meaning. Because of resources disaggregation, a workload is associated to at least one computing node and one memory node. It is therefore possible that it is migrated to a different computing node and keeps using the same memory, or the opposite. We discuss both cases and describe how this can open new opportunities for server consolidation.The last main contribution of this dissertation is related to disaggregated peripherals virtualization. Starting from the assumption that the architecture disaggregation brings many positive effects in general, we explain why it breaks the passthrough peripheral attachment technique (also known as a direct attachment), which is very popular for its near-native performance. To address this limitation we present a design that adapts the passthrough attachment concept to the architecture disaggregation. By this novel design, disaggregated devices can be directly attached to VMs, as if they were plugged locally. Moreover, all modifications do not involve the guest OS itself, for which the setup of the underlying infrastructure is not visible

Capitoli di libri sul tema "Disaggregated Memory":

1

Wang, Tao, Haikun Liu e Hai Jin. "Efficient Remote Memory Paging for Disaggregated Memory Systems". In Algorithms and Architectures for Parallel Processing, 1–20. Cham: Springer Nature Switzerland, 2023. http://dx.doi.org/10.1007/978-3-031-22677-9_1.

Testo completo
Gli stili APA, Harvard, Vancouver, ISO e altri
2

Singhal, Sharad, Clarete R. Crasta, Mashood Abdulla, Faizan Barmawer, Dave Emberson, Ramya Ahobala, Gautham Bhat, Rishi kesh K. Rajak e P. N. Soumya. "OpenFAM: A Library for Programming Disaggregated Memory". In OpenSHMEM and Related Technologies. OpenSHMEM in the Era of Exascale and Smart Networks, 21–38. Cham: Springer International Publishing, 2022. http://dx.doi.org/10.1007/978-3-031-04888-3_2.

Testo completo
Gli stili APA, Harvard, Vancouver, ISO e altri
3

Keeton, Kimberly, Sharad Singhal e Michael Raymond. "The OpenFAM API: A Programming Model for Disaggregated Persistent Memory". In OpenSHMEM and Related Technologies. OpenSHMEM in the Era of Extreme Heterogeneity, 70–89. Cham: Springer International Publishing, 2019. http://dx.doi.org/10.1007/978-3-030-04918-8_5.

Testo completo
Gli stili APA, Harvard, Vancouver, ISO e altri
4

Garrison, John S. "The Pleasure of the Sonnets". In The Pleasures of Memory in Shakespeare's Sonnets, 41–62. Oxford University PressOxford, 2023. http://dx.doi.org/10.1093/oso/9780198857716.003.0003.

Testo completo
Gli stili APA, Harvard, Vancouver, ISO e altri
Abstract (sommario):
Abstract This chapter pursues two interrelated questions: what constitutes pleasure in Shakespeare’s poems and what pleasure does the reader derive from reading them? The discussion utilizes Roland Barthes’s The Pleasure of the Text as a heuristic framework for pursuing these questions, and the chapter takes Sonnet 18 as its central case study, as this poem captures an elevated state of love that is at heart indescribable. The discussion is enhanced by analysis of collective memory and the promise of bliss in Sonnet 55, Sonnet 122, and Sonnet 126. The discussion underscores how memory—both the speaker’s and the reader’s—cannot be disaggregated from the pleasures described in the poems.
5

Grimm, Jannis Julien. "Conclusion and Implications". In Contested Legitimacies. Nieuwe Prinsengracht 89 1018 VR Amsterdam Nederland: Amsterdam University Press, 2022. http://dx.doi.org/10.5117/9789463722650_ch08.

Testo completo
Gli stili APA, Harvard, Vancouver, ISO e altri
Abstract (sommario):
This concluding chapter discusses the findings of the previous chapters critically in light of ongoing theoretical debates on the power of nationalism, the role of affect and emotions, the imprint of violent events such as Rabaa on the collective memory of a people, and the resurgence of autocracy in the Middle East. It highlights the merits of a disaggregated and culture-sensitive interactionist approach to the study of protest and repression, reflects on the implications and limitations of this study, and offers points of contact for further comparative research. Finally, the concluding chapter assesses the prospects for popular resistance in contemporary Egypt.
6

Skrimponis, Panagiotis, Emmanouil Pissadakis, Nikolaos Alachiotis e Dionisios Pnevmatikatos. "Accelerating Binarized Convolutional Neural Networks with Dynamic Partial Reconfiguration on Disaggregated FPGAs". In Parallel Computing: Technology Trends. IOS Press, 2020. http://dx.doi.org/10.3233/apc200099.

Testo completo
Gli stili APA, Harvard, Vancouver, ISO e altri
Abstract (sommario):
Convolutional Neural Networks (CNNs) currently dominate the fields of artificial intelligence and machine learning due to their high accuracy. However, their computational and memory needs intensify with the complexity of the problems they are deployed to address, frequently requiring highly parallel and/or accelerated solutions. Recent advances in machine learning showcased the potential of CNNs with reduced precision, by relying on binarized weights and activations, thereby leading to Binarized Neural Networks (BNNs). Due to the embarassingly parallel and discrete arithmetic nature of the required operations, BNNs fit well to FPGA technology, thus allowing to considerably scale up problem complexity. However, the fixed amount of resources per chip introduces an upper bound on the dimensions of the problems that FPGA-accelerated BNNs can solve. To this end, we explore the potential of remote FPGAs operating in tandem within a disaggregated computing environment to accelerate BNN computations, and exploit dynamic partial reconfiguration (DPR) to boost aggregate system performance. We find that DPR alone boosts throughput performance of a fixed set of BNN accelerators deployed on a remote FPGA by up to 3x in comparison with a static design that deploys the same accelerator cores on a software-programmable FPGA locally. In addition, performance increases linearly with the number of remote devices when inter-FPGA communication is reduced. To exploit DPR on remote FPGAs and reduce communication, we adopt a versatile remote-accelerator deployment framework for disaggregated datacenters, thereby boosting BNN performance with negligible development effort.

Atti di convegni sul tema "Disaggregated Memory":

1

Calciu, Irina, M. Talha Imran, Ivan Puddu, Sanidhya Kashyap, Hasan Al Maruf, Onur Mutlu e Aasheesh Kolli. "Rethinking software runtimes for disaggregated memory". In ASPLOS '21: 26th ACM International Conference on Architectural Support for Programming Languages and Operating Systems. New York, NY, USA: ACM, 2021. http://dx.doi.org/10.1145/3445814.3446713.

Testo completo
Gli stili APA, Harvard, Vancouver, ISO e altri
2

Calabretta, Nicola, Xiaotao Guo, Georgios Exarchakos, Xuwei Xue e Bitao Pan. "Optical Switching for Memory-Disaggregated Datacenters". In Optical Fiber Communication Conference. Washington, D.C.: OSA, 2021. http://dx.doi.org/10.1364/ofc.2021.f2f.4.

Testo completo
Gli stili APA, Harvard, Vancouver, ISO e altri
3

Maruf, Hasan Al, Yuhong Zhong, Hongyi Wang, Mosharaf Chowdhury, Asaf Cidon e Carl Waldspurger. "Memtrade: Marketplace for Disaggregated Memory Clouds". In SIGMETRICS '23: ACM SIGMETRICS International Conference on Measurement and Modeling of Computer Systems. New York, NY, USA: ACM, 2023. http://dx.doi.org/10.1145/3578338.3593553.

Testo completo
Gli stili APA, Harvard, Vancouver, ISO e altri
4

Lim, Kevin, Yoshio Turner, Jose Renato Santos, Alvin AuYoung, Jichuan Chang, Parthasarathy Ranganathan e Thomas F. Wenisch. "System-level implications of disaggregated memory". In 2012 IEEE 18th International Symposium on High Performance Computer Architecture (HPCA). IEEE, 2012. http://dx.doi.org/10.1109/hpca.2012.6168955.

Testo completo
Gli stili APA, Harvard, Vancouver, ISO e altri
5

Gonzalez, Jorge, Alexander Gazman, Maarten Hattink, Mauricio G. Palma, Meisam Bahadori, Ruth Rubio-Noriega, Lois Orosa et al. "Optically Connected Memory for Disaggregated Data Centers". In 2020 IEEE 32nd International Symposium on Computer Architecture and High Performance Computing (SBAC-PAD). IEEE, 2020. http://dx.doi.org/10.1109/sbac-pad49847.2020.00017.

Testo completo
Gli stili APA, Harvard, Vancouver, ISO e altri
6

Zahka, Daniel, e Ada Gavrilovska. "FAM-Graph: Graph Analytics on Disaggregated Memory". In 2022 IEEE International Parallel and Distributed Processing Symposium (IPDPS). IEEE, 2022. http://dx.doi.org/10.1109/ipdps53621.2022.00017.

Testo completo
Gli stili APA, Harvard, Vancouver, ISO e altri
7

Zacarias, Felippe, Paul Carpenter e Vinicius Petrucci. "Dynamic Memory Provisioning on Disaggregated HPC Systems". In SC-W 2023: Workshops of The International Conference on High Performance Computing, Network, Storage, and Analysis. New York, NY, USA: ACM, 2023. http://dx.doi.org/10.1145/3624062.3624174.

Testo completo
Gli stili APA, Harvard, Vancouver, ISO e altri
8

Aguilera, Marcos K., Naama Ben-David, Rachid Guerraoui, Antoine Murat, Athanasios Xygkis e Igor Zablotchi. "uBFT: Microsecond-Scale BFT using Disaggregated Memory". In ASPLOS '23: 28th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 2. New York, NY, USA: ACM, 2023. http://dx.doi.org/10.1145/3575693.3575732.

Testo completo
Gli stili APA, Harvard, Vancouver, ISO e altri
9

Theodoropoulos, Dimitris, Andrea Reale, Dimitris Syrivelis, Maciej Bielski, Nikolaos Alachiotis e Dionisios Pnevmatikatos. "REMAP: Remote mEmory Manager for disAggregated Platforms". In 2018 IEEE 29th International Conference on Application-specific Systems, Architectures and Processors (ASAP). IEEE, 2018. http://dx.doi.org/10.1109/asap.2018.8445095.

Testo completo
Gli stili APA, Harvard, Vancouver, ISO e altri
10

Song, Jongtae, Jiwook Youn, Dae-Ub Kim, Kyeong-Eun Han e Joon Ki Lee. "Performance Evaluation on Optically Disaggregated Memory Architecture". In 2021 International Conference on Information and Communication Technology Convergence (ICTC). IEEE, 2021. http://dx.doi.org/10.1109/ictc52510.2021.9621006.

Testo completo
Gli stili APA, Harvard, Vancouver, ISO e altri

Rapporti di organizzazioni sul tema "Disaggregated Memory":

1

Kommareddy, Vamsee, Clayton Hughes, Simon David Hammond e Amro Awad. Opal: A Centralized Memory Manager for Investigating Disaggregated Memory Systems. Office of Scientific and Technical Information (OSTI), agosto 2018. http://dx.doi.org/10.2172/1467164.

Testo completo
Gli stili APA, Harvard, Vancouver, ISO e altri

Vai alla bibliografia