Research Papers

In this Thought Leadership Paper, Forrester finds that companies lack clear understanding about private clouds and tend to underestimate the key components—storage among them—that are required to achieve maximum benefit from a cloud strategy. This analyst paper takes steps to clarify the private cloud concept and explain the importance of storage to successful cloud computing implementations.

In this Thought Leadership Paper, Forrester finds that companies lack clear understanding about private clouds and tend to underestimate the key components—storage among them—that are required to achieve maximum benefit from a cloud strategy. This analyst paper takes steps to clarify the private cloud concept and explain the importance of storage to successful cloud computing implementations.

This document provides a blueprint for private IaaS implementations on the Microsoft Hyper-V and System Center platform. This document is for customers and partner architects seeking to design private cloud implementations on the NetApp platform.

Reducing or deferring a datacenter’s capital and/or operational expense has become a paramount concern in today’s challenging economic climate. Many storage vendors’ promote functionality and features that can purportedly provide potential cost savings but few actually deliver on this promise. However, one proven feature offered in the open system marketplace since the early 2000’s is thin provisioning.

The statistical bases for current models of RAID reliability are reviewed, and a highly accurate alternative is provided and justified. This new model corrects statistical errors associated with the pervasive assumption that system (RAID group) times-to-failure follow a homogeneous Poisson process, and it corrects errors associated with the assumption that the time-to-failure and time-to-restore distributions are exponentially distributed. Statistical justification for the new model uses theories of reliability of repairable systems. Four critical component distributions are developed from field data. These distributions are for times to catastrophic failure, reconstruction and restoration, read errors, and disk data scrubs. Model results have been verified to predict between 2 and 1,500 times as many double disk failures as estimates made using the mean time-to-data-loss (MTTDL) method. Model results are compared to system-level field data for a RAID group of 14 drives and show excellent correlation and greater accuracy than either MTTDL or Markov models.

In this paper we present the analysis of two large-scale network file system workloads. We measured CIFS traffic for two enterprise-class file servers deployed in the NetApp data center for a three month period. One file server was used by marketing, sales, and finance departments and the other by the engineering department. Together these systems represent over 22TB of storage used by over 1500 employees, making this the first ever large-scale study of the CIFS protocol.

A collection of the latest NetApp Research Papers bound in a single volume.

This article presents an automated technique for visualizing large software architectures using multiple graphical representations, including multi-dimensional scaling, 2-D grid, and spiral layouts. We describe how our software visualization methods were applied to the Network Appliance operating system known as Data ONTAP 7G (ONTAP). We show how each method can be applied to comprehend a specific aspect of ONTAP. This approach can be used by software engineers, architects, and developers to better understand the architecture of their code.

This report analyzes the failure characteristics of storage subsystems. More specifically, we analyzed the NetApp AutoSupport logs collected from about 39,000 storage systems commercially deployed at various customer sites. The data set covers a period of 44 months and includes about 1,800,000 disks hosted in about 155,000 storage shelf enclosures.

In this paper, we present the design and implementation of SWEEPER architecture and backup copy selection algorithms that specifically tackle the problem of quickly and systematically identifying a good recovery point.