TOLLY. No July Standard file open of a 4MB Excel spreadsheet over a Mbps link (100 ms latency) 29X (3 sec) 9X (10 sec) (497 sec)

Similar documents
TOLLY. No October Standard file open of a 4MB Excel spreadsheet over a Mbps link (100 ms latency) 29X (3 sec) 9X (10 sec)

TOLLY. Citrix Systems, Inc. Citrix Systems, Inc. commissioned. Citrix Presentation Server 4.0. Test Summary

TOLLY. No March Fortress Technologies, Inc.

TOLLY. No November 2005 Nortel Ethernet Routing Switch 5510, 5520, 5530 Layer 2 Performance, Resiliency and Ease of Use

TOLLY. VPN Gateway 3070 SSL VPN Throughput, Scalability and Voice Quality Benchmark Evaluation. Test Summary. Test Highlights

TOLLY. Tasman Networks, Inc. commissioned. Tasman Networks, Inc and 1004 Routers. Test Summary

TOLLY. Nortel, Inc. Ethernet Routing Switch 5000 Series. Test Summary

TOLLY. No July 2002

TOLLY. Dell Computer Corp. commissioned. Test Summary. Test Highlights

TOLLY. No June Reflection for the Web v versus IBM SecureWay Host On-Demand v Web-to-Host Software Competitive Evaluation

TOLLY. Nortel Networks. Contivity Extranet Switch Test Summary. Fast Ethernet-to-Fast Ethernet Layer 2 Tunneling Protocol Throughput

2 System boot time improvement up to 38%

T H E TOLLY. No September 2002

TOLLY. Radware, Inc. Radware, Inc. commissioned. DefensePro Test Summary. Throughput Benchmark and Attack Mitigation Evaluation.

T H E TOLLY. No March StreamGroomer Module 200 Flow Regulator and StreamGroomer Manager (SGM) Transactions per second

TOLLY. Extreme Networks, Inc.

TOLLY. No June Functional Verification and Performance Evaluation. Windows 2000 with IIS5 1 Web Server Performance.

Cisco Wide Area Application Services (WAAS) Mobile

Acceleration Performance Tests for IBM Rational ClearTeam Explorer

TOLLY. No August 2002

Symantec Network Access Control Starter Edition

Symantec Network Access Control Starter Edition

T H E TOLLY. No September Dell PowerConnect 3348 vs. 3Com SuperStack 3

WINNER 2007 WINNER 2008 WINNER 2009 WINNER 2010

TOLLY No September 1999

Force10 Networks, Inc.

Features. HDX WAN optimization. QoS

TOLLY. No December 2001 Fujitsu, Ltd. GeoStream R940 IP Switching Node Performance Evaluation. Cause

TOLLY. No July 2001 Nortel Networks, Inc. Contivity 2600 VPN Switch Firewall/VPN Multi-Service Performance Evaluation. Throughput (Mbit/s)

TOLLY. No July 1999

N-Series SoC Based Thin Clients

Performance and Scalability Benchmark: Siebel CRM Release 7 on HP-UX Servers and Oracle9i Database. An Oracle White Paper Released October 2003

WHITE PAPER: BEST PRACTICES. Sizing and Scalability Recommendations for Symantec Endpoint Protection. Symantec Enterprise Security Solutions Group

Symantec Network Access Control Starter Edition

THE TOLLY REPORT. With the advent of the smart grid many utilities will select the technology first and then the meter

Acceleration Performance Tests for IBM Rational ClearCase Remote Client (CCRC)

Benefits of deploying Dell SonicWALL WAN Acceleration with Dell EqualLogic storage

Retired. For more information on HP's ProLiant Security Server visit:

Cabletron Systems Inc. SmartSwitch 9000 Token Ring Switching in Heterogeneous Networks

Citrix CloudBridge Product Overview

QuickSpecs. Available Packs and Purchase Information. ProLiant Essentials Vulnerability and Patch Management Pack v2.1. Overview.

WX CENTRAL MANAGEMENT SYSTEM

Data Sheet: Endpoint Security Symantec Network Access Control Starter Edition Simplified endpoint enforcement

T H E TOLLY G R O U P. Competitive Performance Evaluation versus Cisco Catalyst 6509 Switch. Test Highlights (++, .-, -+, E*'7D1&A -()'7D1&A )-,

Protect enterprise data, achieve long-term data retention

New Cisco 2800 And 3800 Series Integrated Services Router Wan Optimization Bundles

How to deploy a Microsoft Windows 10 image to an AMD processor-based laptop or desktop

HPE ProLiant ML110 Gen10 Server

TOLLY. No March NetScreen Technologies, Inc. 2,000 1,750 1,500 1,250 1, Aggregate throughput.

Performance of Virtual Desktops in a VMware Infrastructure 3 Environment VMware ESX 3.5 Update 2

GIS Application Delivery Challenges & Solutions During State of Emergencies

Cisco WAAS Mobile User Guide

HP Dynamic Deduplication achieving a 50:1 ratio

March MomentumPro V3.1 Technical Specification Guidelines

Microsoft Office SharePoint Server 2007

AMD: WebBench Virtualization Performance Study

VIRTUALIZED BACKUP ADVANTAGES OF THE DELL POWERVAULT DL BACKUP TO DISK APPLIANCE POWERED BY SYMANTEC BACKUP EXEC 2010 R3

McAfee epolicy Orchestrator 4.5 Hardware Sizing and Bandwidth Usage Guide

Contents Overview of the Compression Server White Paper... 5 Business Problem... 7

Boosting Server-to-Server Gigabit Throughput with Jumbo Frames

HPE ProLiant ML110 Gen P 8GB-R S100i 4LFF NHP SATA 350W PS DVD Entry Server/TV (P )

Is BranchCache right for remote, serverless software distribution?

Optimizing NetApp SnapMirror Data Replication with F5 BIG-IP WAN Optimization Manager

XLmanage Version 2.4. Installation Guide. ClearCube Technology, Inc.

T E C H N I C A L S A L E S S O L U T I O N S

NOTE: For more information on HP's offering of Windows Server 2012 R2 products, go to:

A PRINCIPLED TECHNOLOGIES TEST REPORT DELL ACTIVE SYSTEM 800 WITH DELL OPENMANAGE POWER CENTER

HP ProCurve Manager Plus

Assessing performance in HP LeftHand SANs

PMS 138 C Moto Black spine width spine width 100% 100%

Network Security Platform 8.1

Pro: Windows Server 2008 R2, Virtualization Administrator

Oracle Database Appliance X6-2S / X6-2M ORACLE ENGINEERED SYSTEMS NOW WITHIN REACH FOR EVERY ORGANIZATION

NetAlly. Application Advisor. Distributed Sites and Applications. Monitor and troubleshoot end user application experience.

EMC Backup and Recovery for Microsoft SQL Server

TOLLY No April Com Corporation

Consolidation Assessment Final Report

HPE ProLiant DL360 Gen P 16GB-R P408i-a 8SFF 500W PS Performance Server (P06453-B21)

McAfee Network Security Platform

An HP EliteDesk 800 G4 with Intel Optane memory outperformed three configurations with twice as much RAM (Summary report)

Acceleration Systems Technical Overview. September 2014, v1.4

Symantec Backup Exec System Recovery

Partner Pre-Install Checklist: Common Service Platform Collector (CSP-C) for Smart Portal 0.5

Cisco Wide Area Application Services: Secure, Scalable, and Simple Central Management

NX 9500 INTEGRATED SERVICES PLATFORM SERIES FOR THE PRIVATE CLOUD

CSince XenDesktop 4 Enterprise and Platinum editions offer many other

CIFS Acceleration Techniques

EMC Backup and Recovery for Microsoft Exchange 2007 SP1. Enabled by EMC CLARiiON CX4-120, Replication Manager, and VMware ESX Server 3.

System Requirements E 23 rd, Hutchinson KS (866)

MyCloud Computing Business computing in the cloud, ready to go in minutes

WX Client. Product Description. Product Overview DATASHEET

Branch Repeater :51:35 UTC Citrix Systems, Inc. All rights reserved. Terms of Use Trademarks Privacy Statement

IBM TotalStorage SAN Switch F32

TOLLY. No January Throughput (transactions per second) 9,000 8,121 8,160 8,000 7,000 6,000 5,000 4,000 3,000 2,000 1,000

Sage ERP MAS 200 SQL Version 4.50 Supported Platform Matrix Revised as of July 24, 2014

A High-Performance Storage and Ultra- High-Speed File Transfer Solution for Collaborative Life Sciences Research

HPE ProLiant DL580 Gen10 Server

HPE VMware ESXi and vsphere 5.x, 6.x and Updates Getting Started Guide

HP VMware ESXi and vsphere 5.x and Updates Getting Started Guide

Virtual WAN Optimization Controllers

Transcription:

, Inc. ishared Server and ishared Remote Appliances Performance and Functionality Evaluation Premise: Branch-office workers need real-time access to corporate files, services and applications, all while enjoying optimal performance without compromising security. Users need a comprehensive, multi-layered solution that dramatically reduces WAN latency while also creating a coherent, global, file-sharing environment., Inc. commissioned The Tolly Group to evaluate the performance and verify key functionality for the company's ishared Remote and ishared Server appliances running its full suite of ishared IT Services. 's ishared product line consists of network appliances designed to deliver wide area file services (WAFS) and other branchoffice IT services on a single consolidated platform. The ishared Remote appliance is placed at a corporate branch office and allows remote users to access centrally located file and E-mail servers as if they were in the same physical location. The ishared Server appliance is located at the corporate data center, and brokers communications between a network of ishared Remotes. The ishared Remote appliance also delivers a suite of centrally-managed, branch office IT services that includes: E-mail (Exchange), print, Web caching, networking and management services. Tolly Group engineers conducted extensive performance tests designed to measure the effectiveness of the appliances in terms of the response-time improvements T H E TOLLY G R O U P. 206139 July 2006 Test Highlights Accelerates CIFS traffic dramatically, delivering up to 357X improvement in response times over the WAN Accelerates MAPI traffic by optimizing transfer of E-mail over the WAN, delivering a hefty 52X improvement in response times, and up to a 99.6% reduction in bandwidth consumption Delivers end-to-end support for SMB packet signing, allowing operators to maintain both maximum file access performance and security in their networks Delivers IT Services like print, DNS, DHCP and Web caching, resulting in significant savings on capital expenditures for remote offices Allows transparent provisioning of WAFS services during WAN disruption, ensuring data integrity and protecting user data Standard file open of a 4MB Excel spreadsheet over a 1.544- Mbps link (87 sec) 9X (10 sec) 29X (3 sec) Standard file open of a 4MB Excel spreadsheet over a 512- Kbps link 2006 The Tolly Group Page 1 (497 sec) 20X (25 sec) 62X (8 sec) (1 sec) Standard file open of a 4MB PowerPoint file over a 1.544- Mbps link (154 sec) 9X (18 sec) 154X Test Summary Application Acceleration Response Time Improvement Benefit of ishared Remote When Opening Excel and PowerPoint Files Over 512-Kbps and T-1 Links (357 sec) 357X (1 sec) Standard file open of a 4MB PowerPoint file over a 512- Kbps link 16X (22 sec) Source: The Tolly Group, September 2005 Figure 1

delivered to users and the yield in bandwidth reduction from compression and caching of file and E-mail traffic. Tolly Group engineers also verified the presence of several key features and services, including, Server Message Block (SMB) packet signature compatibility, WAFS security and resiliency in the face of WAN circuit disruptions, and the presence of print, DNS, DHCP and Web caching services. Tests were conducted by Tolly Group engineers in September 2005 at facilities in South Plainfield, NJ. Results Response-Time Improvement for WAFS Tolly Group engineers measured the response times of a remote client requesting common Microsoft Excel and PowerPoint files across simulated 1.544-Mbps (T1) and 512-Kbps links. Application Acceleration Response Time Improvement Benefit of ishared Remote When Saving Modified Excel and PowerPoint Files Over 512-Kbps and T-1 Links Standard file open of a 4MB Excel spreadsheet over a 1.544- Mbps link (172 sec) 57X (3 sec) Standard file open of a 4MB Excel spreadsheet over a 512- Kbps link (338 sec) 68X (5 sec) Results show that 87 seconds were consumed when opening a 4MB Excel file over a simulated T1 connection (with 100ms of latency). Opening that same file over a 512-Kbps link consumed 497 seconds. When engineers added the ishared Remote/ ishared Standard file open of a 4MB PowerPoint file over a 1.544- Mbps link (173 sec) 87X (2 sec) 10 (403 sec) (4 sec) Standard file open of a 4MB PowerPoint file over a 512- Kbps link Source: The Tolly Group, September 2005 Figure 2 Server to the equation, that same 4MB Excel file required just nine seconds to open across the simulated T1, and 25 seconds across the simulated 512-Kbps link, representing a response time improvement of 9X and 20X, respectively. This first access is Impact of ishared on WAFS Bandwidth Consumption Percent reduction in bandwidth 4MB Excel file over a 1.544 Mbps link 87% 100% 4MB PowerPoint file over a 1.544 Mbps link 85% 99% Link scenarios 4MB Excel file over a 512 Kbps link 91% 4MB PowerPoint file over a 512 Kbps link 100% 99% 84% Source: The Tolly Group, September 2005 Figure 3 2006 The Tolly Group Page 2

known as a "cold" access, and 's WAN optimization technologies, including compression and latency reductions, work to provide a dramatic performance benefit even on initial requests. Then, on a second of the file across the ishared solution (a "warm" access), the time required to open the 4MB Excel file shrank even further to three seconds over the simulated T1 link and eight seconds over the simulated 512-Kbps link, both representing response time improvements of 29X and 62X over the file open procedure without the aid of any. Tolly Group engineers ran the same test, this time transferring a 4MB PowerPoint file. Without the aid of the ishared appliance, the file transferred in 154 seconds over a simulated T1 link and 357 seconds over the simulated 512-Kbps link. When the ishared Remote/ ishared Server were added to the test bed, the file access took 18 seconds across the T1 and 22 seconds across the 512-Kbps link. There were even more dramatic results when the file was accessed a second time. When the user requested the file a second time, the PowerPoint took just one second over the T1 and over the 512-Kbps simulated links, representing a response-time improvement of 154X and 357X over the file open procedure without the aid of any. (See Figure 1.) This demonstrates the benefit of the WAFS appliance both on initial file access, and when handling files that are requested on a regular basis. Next, engineers measured the response times of a remote client saving both Excel and PowerPoint files over simulated WAN connections. When engineers saved a 4MB Excel file over the 1.544-Mbps link, the file-save operation took 172 seconds with no applied. Over the same connection, using the same file, the file-save operation took just three seconds when the ishared Remote/ ishared Server appliance was added to the configuration. That represents a 57X responsetime improvement. The same test was run again, over the 512-Kbps link, with similar results. This time, the ishared Remote/ ishared Server appliance resulted in a response-time improvement of 68X (five seconds versus 338 seconds). The same file-save operation was applied for the PowerPoint scenario, as well. Here, again, results show that with no applied, the file-save operation took 173 seconds over the T1 link, but just two seconds when the ishared Remote/ ishared Server appliance was deployed. That represents an 87X improvement in the file-save response time. The improvement jumps to 10 when tested over a 512-Kbps link with 250ms of latency. (See Figure 2.) Bandwidth Savings for WAFS Tests show that the ishared Remote/ ishared Server appliances yield sizable improvement in response time for WAFS operations. But there is a second benefit that comes in the form of bandwidth savings. Tolly Group engineers examined the amount of bandwidth utilized during a first- of files through the ishared Remote appliance. Results show that ishared Remote yielded a reduction of WAN This test summary is a rebrand of document #205133., Inc. acquired Tacit Networks in May 2006., Inc. ishared Server and ishared Remote Functionality and Performance, Inc. ishared Appliance Product Specifications* Core Features CIFS and NFS file via WAFS Microsoft Exchange (MAPI) WAN Web (HTTP) caching and optimization Print, DNS, DHCP, and Domain Controller services Software management and distribution optimization Management features Web-based administration Centralized software upgrades SNMP support System-at-a-glance diagnostic console Bandwidth efficiency charting Product Options 100GB, 300GB and 700GB RAID-backed persistent cache sizes Windows- and Linux-based operating system platforms Choice of hardware appliance, qualified IBM, HP, or Dell servers, or software-based deployment For more information contact:, Inc. 10201 N. De Anza Boulevard Cupertino, CA 95014 United States Phone: (408) 873-4400 or (800) 697-2253 Fax: (408) 873-4410 URL: www.packeteer.com *Vendor-supplied information not verified by The Tolly Group 2006 The Tolly Group Page 3

MS Exchange Acceleration Response Time Improvement Benefit of ishared Exchange Services When Accessing E-mail with Excel and PowerPoint Attachments Over 512-Kbps and T-1 Links E-mail access (4MB PowerPoint file) over a 1.544- Mbps link E-mail attachment (4MB Excel file) over a 1.544- Mbps link (100ms latency) 13X (3 sec) 47X (4 sec) E-mail access (4MB Excel file) over a 512- Kbps link (250ms latency) 52X (4 sec) E-mail access (4MB PowerPoint file) over a 512- Kbps link 9X (5 sec) (45 sec) (39 sec) (186 sec) (209 sec) Source: The Tolly Group, September 2005 Figure 4 bandwidth consumption of between 84% and 91% on a first meaning the solution used up to 91% less bandwidth than WAN access without the solution. (See Figure 3.) A second request for the same file shows that there is a 99% to 100% reduction in bandwidth consumption meaning the data has been cached locally on the ishared Remote and is not taxing costly WAN circuits. In fact, the only time that the ishared Remote reaches back to the ishared Server is when it needs new data that is not locally available. Response-Time Improvement for MS Exchange Attachments Response-time is not limited to accessing and saving files over the wide area. With large attachments consuming the vast majority of storage and network resources in MS Exchange environments, companies also need to improve the response times of accessing E-mail over the WAN. enables that via its ishared Exchange Services. ishared Exchange Services ensures that E-mail attachments are delivered only once per branch office and that only file differences are transmitted via the WAN when attachments are revised and saved. Engineers accessed an E-mail with a 4MB PowerPoint attachment from a remote client to a central Exchange server across the simulated WAN. Without the benefit of, the operation took 45 seconds. With the ishared Exchange Services active, the same operation took just five seconds that's a nine-fold improvement in response time over the no scenario. When engineers re-ran the test, this time over a 512-Kbps link with 250ms of latency, the operation consumed 209 seconds when no was used, versus just four seconds with ishared Exchange Services in the configuration. That represents a 52X improvement in response time. (See Figure 4.) Engineers ran the test again, this time substituting a 4MB Excel file. Over the T1 link, the file-save operation took 39 seconds without versus three seconds with ishared Exchange Services a 13X response-time improvement. When engineers ran the test over the 512-Kbps link, the ishared Exchange Services delivered a response time improvement of 47X. Bandwidth Savings for Exchange The solution reduced bandwidth consumption by as much as 99.6% when opening and saving a 4MB Excel file E-mail attachment across a 512-Kbps link. In a baseline scenario where no tool was employed, the 2006 The Tolly Group Page 4

Reduction in bandwidth consumption (%) 4MB Excel file over a 1.544 Mbps link server sent 7,104,147 bytes, as opposed to only 26,752 bytes when the ishared Server/Remote solution handled the operation. (See Figure 5.) Security and IT Services In addition to the performance testing, Tolly Group engineers conducted an evaluation of some of the security, resiliency, and services features offered by the ishared Server and ishared Remote appliances. Security testing: One of the main security features offered by with its ishared Server and ishared Remote appliances is compatibility with SMB (Server Message Block) packet signatures. By default, Microsoft Windows Server 2003 and Microsoft Windows XP Professional operating systems come with SMB signatures enabled which raises the level of security when sending/ receiving data packets across LAN/WAN networks. This SMB signature reduces the chance that an attacker sniffs the packets and re-sends them with malicious ishared Exchange Services Bandwidth Savings Achieved 4MB PowerPoint file over a 1.544 Mbps link 99% 99% Link scenarios 4MB Excel file over a 512 Kbps link 100% 4MB PowerPoint file over a 512 Kbps link purposes. Our hands-on testing confirms that the appliances support SMB signing, and at the same time does not compromise performance of the WAFS or ishared Exchange solution. For this project, Tolly Group engineers conducted all performance tests described in this document with SMB signing enabled. IT services: Another set of features offered by is ishared IT Services. ishared Remote is the platform for common IT Windows environment services such as network shared printers, DHCP and DNS services, and Web caching (Proxy Server). (See Figure 6.) Tolly Group engineers configured each of the IT services and confirmed that they were fully functional in the remote network. For example, for the Web caching feature, engineers were able to configure the Internet browser Mozilla FireFox (ver. 1.0.6) and browse Web pages from the Web server without any problems. WAN Disruption Testing Tolly Group engineers ran a set of five WAN disruption tests comparing a WAN connection without the benefit of services to a WAN connection with the solution to determine how the ishared Remote appliance behaves whenever there is an outage on the WAN. The first test involved transferring a file from the remote site to the headquarters site. Using WAN connectivity without ishared appliances, engineers took down the WAN connection and noticed the filetransfer process immediately aborted, displaying a "Windows Delayed File Write error." However, when the solution was enabled, engineers did not spot any ishared IT Services Services supported File Exchange Print Web caching DNS DHCP 99% te: Amount of bandwidth saved over the link when saving an E-mail attachment. Source: The Tolly Group, September 2005 Figure 5 Source: The Tolly Group, September 2005 Figure 6 2006 The Tolly Group Page 5

errors or issues during the WAN disruption since the ishared Remote detected the outage and cached the operation. Even though there was no WAN connectivity between the ishared Server and the ishared Remote appliances, the end-user file was stored in the ishared Remote until the WAN connection was restored and then, the file was transmitted to the ishared Server and stored on the headquarters network drive. (See Figure 7.) The second test consisted of breaking the WAN connection while saving a Wordpad document on the headquarters network drive. Engineers verified that the ishared appliances did not disrupt the saving process to the end user, since the changes to the file were stored on the ishared Remote and updated when the connection was restored. However, without the benefit of the ishared, the Wordpad saving process hung and engineers were confronted with an application error message of "not responding." The third test consisted of editing the same Wordpad file that was saved to the backend server previously. Since the file was cached also on the ishared Remote, the changes were captured locally and updated to the backend after the connection was restored. The fourth test verified that the ishared Remote appliance was able to maintain all cached files saved during a WAN outage. Engineers rebooted the appliance and saw that effectively the system maintained all files saved. In addition, the last test showed that those files that were already cached before the WAN disruption were accessible to the end user as "readonly" until the WAN connection was fully restored, enabling users to access critical data even if the WAN is down. Test Configuration and Methodology For performance and validation tests of WAFS and other ishared IT Services, Tolly Group engineers tested the ishared Server and ishared Remote. The server-side and remote-side (ishared Server and ishared Remote) are Microsoft Windows-powered appliances with Microsoft Windows Server 2003 Standard Edition installed and equipped with an Intel Xeon 2.4-GHz processor, 100GB of hard drive space and 1GB of RAM. In addition, both appliances come with 's Web Management Console pre-installed (software ver. 2.4.1.0). Impact of ishared on WAN Disruption Events WAN Disruption Test WAN ( Solution) WAN with Solution "Drag and drop" file copy onto the target network drive. Break the WAN during the file copy. Using Wordpad, save document onto the target network drive. Break the WAN connection during the file save. File copy is interrupted with error message reporting that data may be lost or corrupted. Engineers also verified that file was corrupted on the file server. File save was interrupted with an error message reporting that data may be lost or corrupted. Engineers verified that the file was corrupted on the file server. File copy completes successfully without errors. File save completes successfully without errors. With the WAN down, continue to make edits and save onto the target network drive. Does not apply. (Previous operation did not complete.) File saves continue to complete successfully. With the WAN down, after file copy or save, shutdown the ishared Remote appliance after two minutes and restore the WAN. With the WAN down, attempt to access files from the target network drive. Does not apply. (Initial save/copy failed.) The share/folder cannot be accessed. The copy/save operation is reflected on the file server successfully after the WAN and the appliance power have been restored. Files can still be browsed and opened as "read-only" through the network drive. Source: The Tolly Group, September 2005 Figure 7 2006 The Tolly Group Page 6

Test Bed Diagram ishared Server ishared Remote NETGEAR Model FS605-v2 five-port Fast Ethernet Switch Simulated WAN Soekris Engineering net4801 Linksys five-port workgroup hub ` Dell Dimension 2400 Windows Server 2003 Enterprise Edition - File Server - Mail Server (Microsoft Exchange Server 2003) - Web Server (IIS 6.0) Dell Latitude D600 Windows XP Professional (Client #1) - Ethereal Network Protocol Analyzer (ver. 0.10.12) Hewlett-Packard e-vectra Windows XP Professional (Client #2) Source: The Tolly Group, September 2005 Figure 8 Tolly Group engineers configured an additional Windows Server 2003 (Enterprise Edition) machine as a file server, Web server and E-mail server. The operating system and other applications were set up on a Dell Dimension 2400 equipped with a 2.66-GHz Intel Pentium 4 processor and 256MB of RAM. For the E-mail server side, engineers installed and configured Microsoft Exchange Server 2003 (Service Pack 1), and for the Web server side, engineers configured IIS (Internet Information Services) version 6.0 with default settings. Engineers utilized a Soekris Engineering net4801 (version 1.11) WAN emulation hardware and software switch (with three Fast Ethernet ports available) to provide the test environment with two simulated WAN link speeds tested: T1 (1.544 Mbps) and 512 Kbps. From the client-side, engineers con-figured a Dell Latitude D600 laptop with Windows XP Professional (Service Pack 1) and Microsoft Office XP (Service Pack 2), equipped with a 1.3- GHz Intel Pentium M processor and 512MB of RAM. The second client that engineers configured was a Hewlett-Packard e-vectra desktop with Windows XP Professional and Microsoft Office XP (Service Pack 2), equipped with a 600-MHz Intel Pentium III processor and 128MB of RAM. Two five-port Fast Ethernet hub/ switches (Linksys Workgroup Hub/NETGEAR Model FS605-v2 Fast Ethernet switch) were connected to create two independent networks. (See Figure 8.) The Soekris net4801 controlled bandwidth and latency between these networks. Both the I- shared Server appliance and Windows Server 2003 file and mail server were connected to one network. The ishared Remote appliance and the two Windows XP clients were deployed on the adjacent network on the other side of the WAN emulator and simulated a client connection at various speeds (1.544 Mbps and 512 Kbps). Test scenarios consisted of two performance tests (Response time improvement for file transfer and ishared Exchange Services) and three validation feature tests (WAFS Wide Area File Services security, WAN Disruption and IT Services). For the performance tests, engineers logged both bandwidth utilization and response times across the LAN/WAN infrastructure as two different file types were transmitted Microsoft Excel and Power-Point. Both programs were used since they 2006 The Tolly Group Page 7

incorporate low- and high-bandwidth demands, from a basic calculation and word processing to graphical and image manipulation respectively. Each file was 4MB. Engineers used a standard stopwatch to take the response times. Ethereal Network Protocol Analyzer test tool (ver. 0.10.12) was used to capture the data and observe the bandwidth utilization across the network. The second pair of tests consisted of validating claims in terms of security and user services for remote networks. Tolly Group engineers tested WAFS performance by opening and saving shared files from a remote network. Using a Windows-based PC, engineers ran five test scenarios: WAN file open, WAN file save, cold file open, warm file open, and file save. The cold open file represented the first of a file through the appliances, while the warm open file represents the second of a file through the appliances. Since the file has already been cached locally on the ishared Remote, it further reduces the amount of time required for the user to access the file. The Soekris WAN emulator was used to throttle a Fast Ethernet network up to speeds of T1 (1.544 Mbps) and 512 Kbps with roundtrip latencies of 100ms and 250ms respectively. For the ishared Exchange Services test, engineers sent an E- mail using Microsoft Outlook from client #1 to client #2 and recorded the time necessary for client #2 to save a 4MB file attachment (both PowerPoint and Excel files). Lastly, engineers tested and validated the WAFS security by checking SMB signatures and the IT Services offered by the ishared Remote appliance such as print services, DNS, DHCP and Web caching. The Tolly Group gratefully acknowledges the providers of test equipment used in this project. Vendor Product Web address Soekris Engineering, Inc. net4801 Switch ver. 1.1 http://www.soekris.com/net4801.htm Public Domain Ethereal Analyzer ver. 0.10.12 http://www.ethereal.com Terms of Usage USE THIS DOCUMENT ONLY IF YOU AGREE TO THE TERMS LISTED HEREIN. This document is provided, free-of-charge, to help you understand whether a given product, technology or service merits additional investigation for your particular needs. Any decision to purchase must be based on your own assessment of suitability. This evaluation was focused on illustrating specific features and/or performance of the product(s) and was conducted under controlled, laboratory conditions and certain tests may have been tailored to reflect performance under ideal conditions; performance may vary under real-world conditions. Users should run tests based on their own real-world scenarios to validate performance for their own networks. Commercially reasonable efforts were made to ensure the accuracy of the data contained herein but errors and/or oversights can occur. The test/audit documented herein may also rely on various test tools the accuracy of which is beyond our control. Furthermore, the document relies on certain representations by the sponsor that are beyond our control to verify. Among these is that the software/hardware tested is production or production track and is, or will be, available in equivalent or better form to commercial customers. The Tolly Group provides a fee-based service to assist users in understanding the applicability of a given test scenario to their specific needs. Contact us for information. When foreign translations exist, the English document is considered authoritative. To assure accuracy, only use documents downloaded directly from The Tolly Group's Web site. Sponsor:, Inc Document number: 206139 Product class: Application appliance Products under test: ishared Server ver. 2.4.1.0 ishared Remote ver. 2.4.1.0 Testing window: September 2005 Project Profile Software status: Generally available For more information on this document, or other services offered by The Tolly Group, visit our World Wide Web site at http://www.tolly.com, send E-mail to sales@tolly.com, call (561) 391-5610. Information technology is an area of rapid growth and constant change. The Tolly Group conducts engineering-caliber testing in an effort to provide the internetworking industry with valuable information on current products and technology. While great care is taken to assure utmost accuracy, mistakes can occur. In no event shall The Tolly Group be liable for damages of any kind including direct, indirect, special, incidental, and consequential damages which may result from the use of information contained in this document. All trademarks are the property of their respective owners. The Tolly Group doc. 206139 rev. clk 11 July 06 2006 The Tolly Group Page 8