Storage Performance in the Cloud
This article describes the problem of storage performance in the cloud and the architecture of Zadara Storage to provide high performance, high reliable storage in the cloud.
It is no secret that one of the main challenges for applications in the cloud is storage performance. The storage performance problem in the cloud has multiple facets. It is not just low vs. high performance, but also the fact that performance varies over time an done customer can impact the performance of others. There is also the issue of storage performance monitoring. How does one know if an application is running slow because of storage performance or because of other factors like networking, IO concurrency, latencies, etc.
The main issue is that none of the current storage systems and software were designed with cloud in mind. Even high-end SAN storage arrays were designed for a single tenant, a single manager, the storage admin who knows what everybody else is doing, the super guru that knows how to interpret iostat! Aside from lack of multi-tenancy isolation, the scalability is not even close to what is needed in the cloud.And worse yet is the inability to accommodate different types of customers with a single system. As a result, cloud providers start with the lowest common denominator.
All these reasons result in a cloud environment into which it is hard to move classical IT applications that need low latency. Instead, other forms of databases and applications that can live with long latencies, low number of IOPS and inconsistencies are the ones that are more popular in the cloud. But the problems are solvable with a new storage architecture. A storage architecture that is not based on“scale-out,” but is designed to host multiple tenants, each with their own needs. In this storage architecture,each user, at least the ones that need consistency and performance, must have dedicated resources. Dedicated drives, dedicated cache, dedicated cores - so that IOPS can be guaranteed as if the user had a SAN array in his or her own data center. And just like in a private environment, each user needs to have access to performance data (for monitoring), and to a management console (for control over and tuning of the environment). This new architecture needs to allow for different amounts of resources to be allocated to different users,depending on their needs.
And this architecture needs to allow metering of these resources, in order to charge customers according to actual usage.
In summary, a new cloud storage architecture is needed - one that is not based on “scale out,” but it is more similar to the compute cloud.
This is what Zadara Storage has done!. We took the architecture of Cloud Compute and applied to the Cloud Storage, to be able to create a very scalable system, highly reliable where each tenant can control their own storage. In addition to that, we worked in providing resource allocation per customer (i.e CPU, memory, networking and storage drives) to eliminate the noisy neighbor problem. Our Virtual Private Storage Arrays in the cloud, perform and behave like classical SAN/NAS storage in the data center. They have the reliability and features of enterprise storage with the elasticity and pay per use of the cloud.
The above information is edited by 10GTEK.
10GTEK TRANSCEIVERS CO., LTD (Hereinafter refered to as 10GTEK) is specialized in developing and manufacturing Fiber Optical Transceivers and High Performance Cables which are wildly applied in Datacom, Telecom and CATV, providing customers with top quality and cost effective products. Our High Speed Cables cover Passive SFP+ Cable, Active SFP+ Cable, QSFP+ cables, MiniSAS (SFF-8088) Cables, CX4 Cables, Harness cables, Breakout Cables, Patchcords. We also manufacture Fiber Optic Transceivers like 10G XFP, 10G SFP+, SFP DWDM/ CWDM, GBIC, etc. The prompt response and excellent customer support contribute to clients‘ full satisfaction.Today, 10GTEK has been growing fast in the optical field for its unique and competitve excellence which has got a high attention from datacom and telecom.
This article reader also like: Little data, big data and very big data (VBD) or big BS?