Elasticity is a fundamental property in cloud computing that has recently witnessed major developments This article reviews both classical and recent elasticity solutions and provides an overview of containerization, a new technological trend in lightweight virtualization It also discusses major issues and research challenges related to elasticity in cloud computing We comprehensively review and analyze the proposals developed in this field We provide a taxonomy of elasticity mechanisms according to the identified works and key properties Compared to other works in literature, this article presents a broader and detailed analysis of elasticity approaches and is considered as the first survey addressing the elasticity of containers

/pdf/elasticity-in-cloud-computing-state-of-the-art-and-research-32fme9yoqc.pdf

Elasticity in Cloud Computing: State of the Art and Research Challenges

Applications in virtualized data centers are often subject to Service Level Objectives (SLOs) regarding their performance (e.g., latency or throughput). In order to fulfill these SLOs, it is necessary to allocate sufficient resources of different types (CPU, memory, I/O, etc.) to an application. However, the relationship between the application performance and the resource allocation is complex and depends on multiple factors including application architecture, system configuration, and workload demands. In this paper, we present a model-based approach to ensure that the application performance meets the user-defined SLO efficiently by runtime "vertical scaling" (i.e., adding or removing resources) of individual virtual machines (VMs) running the application. A layered performance model describing the relationship between the resource allocation and the observed application performance is automatically extracted and updated online using resource demand estimation techniques. Such a model is then used in a feedback controller to dynamically adapt the number of virtual CPUs of individual VMs. We have implemented the controller on top of the VMware vSphere platform and evaluated it in a case study using a real-world email and groupware server. The experimental results show that our approach allows the managed application to achieve SLO satisfaction in spite of workload demand variation while avoiding oscillations commonly observed with state-of-the-art threshold-based controllers.

Runtime Vertical Scaling of Virtualized Applications via Online Model Estimation

Concepts and technologies are disclosed herein for providing a network virtualization policy management system. An event relating to a service can be detected, and virtual machines and virtual network functions that provide the service can be identified. A first policy that defines allocation of hardware resources to host the virtual machines and the virtual network functions can be obtained, as can a second policy that defines deployment of the virtual machines and the virtual network functions to the hardware resources. The hardware resources can be allocated based upon the first policy and the virtual machines and the virtual network functions can be deployed to the hardware resources based upon the second policy.

Network Virtualization Policy Management System

Resource provisioning in cloud computing is typically coarse-grained. For example, entire CPU cores may be allocated for periods of up to an hour. The Resource-as-a-Service cloud concept has been introduced to improve the efficiency of resource utilization in clouds. In this concept, resources are allocated in terms of CPU core fractions, with granularities of seconds. Such infrastructures could be created using existing technologies such as lightweight virtualization using LXC or by exploiting the Xen hyper visor's capacity for vertical elasticity. However, performance models for determining how much capacity to allocate to each application are currently lacking. To address this deficit, we evaluate two performance models for predicting mean response times: the previously proposed queue length model and the novel inverse model. The models are evaluated using 3 applications under both open and closed system models. The inverse model reacted rapidly and remained stable even with targets as low as 0.5 seconds.

/pdf/towards-faster-response-time-models-for-vertical-elasticity-4lokb68p9b.pdf

Towards Faster Response Time Models for Vertical Elasticity

When designing modern software, care must be taken to allow for applications to scale based on the demands of its users while still accommodating flexibility in development. Recently, microservices architectures have garnered the attention of many organizations—providing higher levels of scalability, availability, and fault isolation. Many organizations choose to host their microservices architectures in cloud data centres to offset costs. Incidentally, data centres become over-encumbered during peak usage hours and underutilized during off-peak hours. Traditional microservice scaling methods perform either horizontal or vertical scaling exclusively. When used in combination, however, these methods offer complementary benefits and compensate for each other's deficiencies. To leverage the high availability of horizontal scaling and the fine-grained resource control of vertical scaling, we developed two novel hybrid autoscaling algorithms and a dedicated network scaling algorithm and benchmarked them against Google's popular Kubernetes horizontal autoscaling algorithm. Results indicated up to 1.49x speedups in response times for our hybrid algorithms, and 1.69x speedups for our network algorithm under high-burst network loads.

HyScale: Hybrid and Network Scaling of Dockerized Microservices in Cloud Data Centres

Most modern hypervisors offer powerful resource control primitives such as reservations, limits, and shares for individual virtual machines (VMs). These primitives provide a means to dynamic vertical scaling of VMs in order for the virtual applications to meet their respective service level objectives (SLOs). VMware DRS offers an additional resource abstraction of a resource pool (RP) as a logical container representing an aggregate resource allocation for a collection of VMs. In spite of the abundant research on translating application performance goals to resource requirements, the implementation of VM vertical scaling techniques in commercial products remains limited. In addition, no prior research has studied automatic adjustment of resource control settings at the resource pool level. In this paper, we present AppRM, a tool that automatically sets resource controls for both virtual machines and resource pools to meet application SLOs. AppRM contains a hierarchy of virtual application managers and resource pool managers. At the application level, AppRM translates performance objectives into the appropriate resource control settings for the individual VMs running that application. At the resource pool level, AppRM ensures that all important applications within the resource pool can meet their performance targets by adjusting controls at the resource pool level. Experimental results under a variety of dynamically changing workloads composed by multi-tiered applications demonstrate the effectiveness of AppRM. In all cases, AppRM is able to deliver application performance satisfaction without manual intervention.

Application-driven dynamic vertical scaling of virtual machines in resource pools

A system and method for managing resources in a distributed computer system that includes at least one resource pool for a set of virtual machines (VMs) utilizes a set of desired individual VM-level resource settings that corresponds to target resource allocations for observed performance of an application running in the distributed computer system. The set of desired individual VM-level resource settings are determined by constructing a model for the observed application performance as a function of current VM-level resource allocations and then inverting the function to compute the target resource allocations in order to meet at least one user-defined service level objective (SLO). The set of desired individual VM-level resource settings are used to determine final RP-level resource settings for a resource pool to which the application belongs and final VM-level resource settings for the VMs running under the resource pool, which are then selectively applied.

Performance-driven resource management in a distributed computer system

A cloud management server and method for performing automatic placement of clients in a distributed computer system uses a list of compatible clusters to select an affinity cluster to place the clients associated with an affinity constraint. As part of the placement method, a cluster that cannot satisfy any anti-affinity constraint associated with the clients and the affinity constrain is removed from the list of compatible clusters. After the affinity cluster has been selected, at least one cluster in the distributed computer system is also selected to place clients associated with an anti-affinity constraint.

Automatic placement of clients in a distributed computer system satisfying constraints

A system and method for performing automatic remediation in a distributed computer system with multiple clusters of host computers uses the same placement selection algorithm for initial placements and for remediation placements of clients. The placement selection algorithm is executed to generate a placement solution when a remediation request in response to a remediation-requiring condition in the distributed computer system for at least one client running in one of the multiple clusters of host computers is detected and a remediation placement problem for the client is constructed. The placement solution is then implemented for the client for remediation.

Automatic remediation in a distributed computer system with multiple clusters of host computers

Techniques for enabling virtual machine (VM) recovery on non-shared storage in a single virtual infrastructure management server (VIMS) instance are provided. In one set of embodiments, a VIMS instance can receive an indication that a VM in a first cluster of the VIMS instance has failed, and can determine whether the VM's files were being replicated to a storage component of the VIMS instance at the time of the VM's failure. If the VM's files were being replicated at the time of the failure, the VIMS instance can search for and identify a cluster of the VIMS instance and a host system within the cluster that (1) are compatible with the VM, and (2) have access to the storage component. The VIMS instance can then cause the VM to be restarted on the identified host system of the identified cluster.

Parth Shah

Papers

Application-driven dynamic vertical scaling of virtual machines in resource pools

Performance-driven resource management in a distributed computer system

Automatic placement of clients in a distributed computer system satisfying constraints

Automatic remediation in a distributed computer system with multiple clusters of host computers

Virtual machine recovery on non-shared storage in a single virtual infrastructure management instance