ZIVM: A Zero-Copy Inter-VM Communication Mechanism for Cloud Computing

Abstract

With the advent of virtualization technology and its propagation to the infrastructure of Cloud distributed systems, there is an emergent request for more effective means of communication between virtual machines (VMs) lying on distributed memory than traditional message based communication means. This paper presents a distributed virtual shared memory mechanism called ZIVM (Zero-copy Inter-VM) to ease the programming of inter-VM communications in a transparent way and at the same time to achieve comparable and bearable execution performance. ZIVM has been implemented as a virtual cluster on a hosted virtual machine using the KVM hypervisor. Experimental results have shown near native performance in terms of latency and bandwidth.

Naplus: a software distributed shared memory for virtual clusters in the cloud

“…Recently, Cam Macdonell proposed a new mechanism, called Nahanni for sharing memory between VMs on the same physical host. A similar idea is also exhibited in ZIVM , a zero‐copy inter‐VM communication mechanism for cloud computing. Naplus is based on Nahanni and extends it with the lock‐based protocol developed in JIAJIA to enable the DSM model.…”

“…The approaches are usually implemented and discussed in the virtualization environments KVM and Xen. For example, most approaches are implemented in Xen, such as while some others are discussed in KVM.…”

On-Chip Message Passing Sub-System for Embedded Inter-Domain Communication

“…Inter-domain refers to different virtual partitions running on the same physical system [7]. IDC is typically performed by either shared memory, which requires complex memory mappings, or Hypervisormediated message passing [4] which simplifies memory management, but requires an expensive transition to the Hypervisor and back [14].…”

“…Hypervisor's IDC software support is simplified, enabling functionality to focus on hardware virtualization and partition scheduling. The main contribution of this letter is a hardware mechanism which simplifies required software management and associated overhead compared to [4], [14], [15], [2], suitable for the resource-constrained embedded domain unlike other state of the art mechanisms [8], [9], [10].…”

Τεχνικές Αποδοτικής Εικονικοποίησης Επιταχυντών Για Νεφοϋπολογιστικά Περιβάλλοντα

“…Αντίστοιχες προσεγγίσεις με τις προαναφερθείσες εργασίες έχουν προταθεί επίσης και για άλλους hypervisors [14,28]. Πιο συγκεκριμένα, στο [14] οι Diakhate et al έχουν προτείνει μία προσέγγιση για το KVM με χρήση τεχνικών κοινής μνήμης μέσω μίας εικονικής συσκευής ανταλλαγής μηνυμάτων στοχεύοντας εικονικοποιημένα πλαίσια εφαρμογών MPI.…”