Adding flexibility to multi-tenant networks

Georgios Ioannidis
EPFL, 2018
Thèse EPFL

Résumé

Cloud computing has been experiencing sharp development over the last years, leading to an increased demand for application migration to the cloud. Cloud providers, in an effort to attract more customers and earn their confidence, offer to tenants the illusion of an isolated network, exposing familiar abstractions. At the same time, creating this illusion poses challenging problems for the providers, as one tenant's traffic may interfere with another's in complicated, unpredictable ways.

First, new challenges have arisen in administering access-control rules (ACLs). On the one hand, installing ACLs at the server is incompatible with bare-metal support and introduces unnecessary performance overhead. On the other hand, offloading the most popular ACLs on the limited hardware memory in Top-of-Rack (ToR) switches should not be conducted naïvely, as the existence of wildcard rules presents inter-rule dependencies that must be respected.

Second, tenants' demands have evolved beyond requesting hardware resources; for instance, tenants may require bandwidth provisions between their resources or optimized access to a specific cloud service, e.g., a Mail server or a Database. Cloud providers have not adequately adapted to these expanding demands, therefore elevating hardware resources to "first class citizens," as non-hardware constraints are not considered during resource allocation, instead they are applied afterwards.

In this thesis we propose two architectures that facilitate cloud providers in managing their shared network resources in a flexible way. First, we demonstrate virtual flow tables, a ToR architecture that handles ACLs using a two-level memory hierarchy. The most popular ACLs are stored in the limited hardware memory, respecting any dependencies between wildcard rules, while the ToR's supervisor engine maintains access to the entire ACL rule-set. Second, we present a two-tiered architecture for scheduling cloud resources, consisting of a resource-agnostic scheduling layer and a resource-specific enforcement layer. Network resources and constraints are taken into consideration during resource scheduling, instead of afterwards, while resource provisioning, as well as general network-management policies, are delegated to the resource-specific tier.

Official source

https://infoscience.epfl.ch/record/234534?ln=fr

À propos de ce résultat

Cette page est générée automatiquement et peut contenir des informations qui ne sont pas correctes, complètes, à jour ou pertinentes par rapport à votre recherche. Il en va de même pour toutes les autres pages de ce site. Veillez à vérifier les informations auprès des sources officielles de l'EPFL.

Concepts associés

Chargement

Publications associées

Chargement

Georgios Ioannidis
EPFL, 2018
Thèse EPFL

Résumé

Official source

https://infoscience.epfl.ch/record/234534?ln=fr

À propos de ce résultat

Concepts associés (23)

Le cloud computing , en français l'informatique en nuage (ou encore l'infonuagique au Canada), est la pratique consistant à utiliser des serveurs informatiques à distance et hébergé

In computer networking, a bare-metal server is a physical computer server that is used by one consumer, or tenant, only. Each server offered for rental is a distinct physical piece of hardware that

Un matériel informatique (en anglais : hardware) est une pièce ou composant d'un appareil informatique. C'est la partie physique de l’informatique elle est appairée avec le logiciel (software ou fir

Afficher plus

Concepts associés

Chargement

Publications associées (3)

Publications associées

Chargement

Rack-Scale Memory Pooling for Datacenters

Stanko Novakovic

The rise of web-scale services has led to a staggering growth in user data on the Internet. To transform such a vast raw data into valuable information for the user and provide quality assurances, it is important to minimize access latency and enable in-memory processing. For more than a decade, the only practical way to accommodate for ever-growing data in memory has been to scale out server resources, which has led to the emergence of large-scale datacenters and distributed non-relational databases (NoSQL). Such horizontal scaling of resources translates to an increasing number of servers that participate in processing individual user requests. Typically, each user request results in hundreds of independent queries targeting different NoSQL nodes - servers, and the larger the number of servers involved, the higher the fan-out. To complete a single user request, all of the queries associated with that request have to complete first, and thus, the slowest query determines the completion time. Because of skewed popularity distributions and resource contention, the more servers we have, the harder it is to achieve high throughput and facilitate server utilization, without violating service level objectives. This thesis proposes rack-scale memory pooling (RSMP), a new scaling technique for future datacenters that reduces networking overheads and improves the performance of core datacenter software. RSMP is an approach to building larger, rack-scale capacity units for datacenters through specialized fabric interconnects with support for one-sided operations, and using them, in lieu of conventional servers (e.g. 1U), to scale out. We define an RSMP unit to be a server rack connecting 10s to 100s of servers to a secondary network enabling direct, low-latency access to the global memory of the rack. We, then, propose a new RSMP design - Scale-Out NUMA that leverages integration and a NUMA fabric to bridge the gap between local and remote memory to only 5× difference in access latency. Finally, we show how RSMP impacts NoSQL data serving, a key datacenter service used by most web-scale applications today. We show that using fewer larger data shards leads to less load imbalance and higher effective throughput, without violating applications¿ service level objectives. For example, by using Scale-Out NUMA, RSMP improves the throughput of a key-value store up to 8.2× over a traditional scale-out deployment.

EPFL2017

Chargement

In large scale data centers, network infrastructure is becoming a major cost component; as a result, operators are trying to reduce expenses, and in particular lower the amount of hardware needed to achieve their performance goals (or to improve the performance achieved for a given amount of hardware). In this thesis, we explore the use of locality in the data center to meet these demands. In particular, we leverage locality in two different projects: (1) VNToR, which moves network virtualization from the server to the top-of-rack (ToR) switch, thereby reducing the server hardware needed to achieve a certain performance, and (2) Criss-Cross, which makes the network topology reconfigurable, thereby reducing the network hardware needed to switch typical data center workloads with a given level of performance. VNToR exploits the locality of traffic flows as well as their long-tailed behavior in the design of a virtual flow table, which extends the hardware flow table of off-the-shelf top-of-rack switches. VNToR uses this virtual flow table in a hybrid data plane that consists of both a hardware as well as a software data plane. This way it can (1) store tens or even hundreds of thousands of access rules, (2) adapt to traffic-pattern changes, typically in less than one millisecond, and (3) uses only commodity switching hardware with a minimal amount of data path memory (4)~without compromising latency or throughput. Criss-Cross is a hierarchical, reconfigurable topology for large-scale data centers. The locality in rack-level flows allows Criss-Cross to adjust its topology to the current traffic patterns. We show that Criss-Cross preserves many of the advantages of Clos topologies: (1) it maintains their hierarchy, (2) the simple routing algorithms, (3) their regular layout of connections for simple physical deployability, and (4) the compatibility to existing management approaches. We demonstrate that for a group-based communication pattern, Criss-Cross improves the average flow completion time by 5.5x and the 99th percentile by 6.3x. For a purely random point-to-point traffic pattern, it improves the flow completion time by 2.2x on average and 3x at the 99th percentile.

EPFL2019

Chargement

Understanding and Mitigating Latency Variability of Latency-Critical Applications

Mia Primorac

A major theme of IT in the past decade has been the shift from on-premise hardware to cloud computing. Running a service in the public cloud is practical, because a large number of resources can be bought on-demand, but this shift comes with its own set of challenges, e.g., customers having less control over their environment. In scale-out deployments of on-line data-intensive services, each client request typically executes queries on many servers in parallel, and its final response time is often lower-bounded by the slowest query execution. There are numerous hard-to-trace reasons why some queries finish later than others. Latency variability appears at different timescales. To understand its cause and impact we need to measure it correctly. It is clear that we need hardware support to measure latencies on the nanosecond-scale, and that software-based tools are good-enough for the millisecond- scale, but it is not clear whether software-based tools are also reliable at the microsecond-scale, which is of growing interest to a large research and industry community. Measuring is a first step towards understanding the problems that cause latency variability, but what if some of them are extremely complex, or fixing them is out of reach given a service providerâs restrictions? Even large companies that own datacenters and build their own software and hardware stack sometimes suffer from hard-to-understand or hard-to-fix performance issues. Medium-sized companies that simply use shared public infrastructure, and do not themselves develop most of the code running on the machines, have limited capabilities and often cannot rule out each and every source of system interference. This reality inspired the line of research on what is known as hedging policies. By using redundant requests we can reduce overall latency at the cost of consuming additional resources. This dissertation characterizes latency variability of interactive services, shows how to measure it and how to mitigate its effect on end-to-end latency without sacrificing system capacity. Concretely, this dissertation makes three contributions: First, it shows how to measure microsecond-scale latency variability with both low cost and high precision, with the help of kernel bypass and hardware timestamps. Second, it empirically derives a lower bound on the tail latency that any implementable hedging policy might achieve for a given workload. Through evaluating our lower bound on a large parameter space, we determine when hedging is beneficial. Lastly, we describe and evaluate a practical policy, LAEDGE, that approximates our theoretical lower bound and achieves as much as half of its hedging potential. We show the applicability of our solution to real applications deployed in the public cloud where LAEDGE outperforms the state-of-the-art scheduling policies by up to 49%, averaged on low to medium load.

EPFL2020

Chargement

Rack-Scale Memory Pooling for Datacenters

Stanko Novakovic

EPFL2017

EPFL2019

Understanding and Mitigating Latency Variability of Latency-Critical Applications

Mia Primorac

EPFL2020