Broadcom Inc.
Emulex® Gen 7 LPe35002 Host Bus Adapter
NVMe over Fibre Channel vs. SCSI Performance in VMware ESXi 7 Environments
Document number:
220117
Release Date:
01 May 2020
Innovations in server and storage technologies such as all-flash arrays, NVMe, and NVMe over Fabrics (NVMe-oF) are delivering record speeds and lower latency to improve application performance. As a result, enterprise datacenters require high performance Gen 7 32G Fibre Channel (32GFC) throughput for server connectivity that also provides extreme reliability and scalability under load.
Fibre Channel leads the industry as the first transport with complete NVMe over Fabrics ecosystem support. Every major server vendor is shipping NVMe standard and offers Emulex NVMe over Fibre Channel (NVMe/FC) capable Host Bus Adapters (HBAs).
Broadcom commissioned Tolly to benchmark the performance of its Emulex Gen 7 LPe35002 32GFC HBAs running NVMe/FC against the performance of the same adapters running in SCSI mode. The series of tests included database TPC-C-like tests using Microsoft SQL Server 2017 and Oracle Database 19c running on Red Hat Enterprise Linux 8 (RHEL 8). The Emulex LPe35002 HBA running in NVMe/FC mode demonstrated superior performance in both test scenarios.
Storage
|
Broadcom Inc.
Emulex® Gen 7 LPe35002 Host Bus Adapter
NVMe over Fibre Channel vs. SCSI Performance
Document number:
220122
Release Date:
01 May 2020
Innovations in server and storage technologies such as all-flash arrays, NVMe, and NVMe over Fabrics (NVMe-oF) are delivering record speeds and lower latency to improve application performance. As a result, enterprise datacenters require high performance Gen 7 32G Fibre Channel (32GFC) throughput for server connectivity that also provides extreme reliability and scalability under load.
Fibre Channel leads the industry as the first transport with complete NVMe over Fabrics ecosystem support. Every major server vendor is shipping NVMe standard and offers Emulex NVMe over Fibre Channel (NVMe/FC) capable Host Bus Adapters (HBAs).
Broadcom commissioned Tolly to benchmark the performance of its Emulex Gen 7 LPe35002 32GFC HBAs running NVMe/FC against the performance of the same adapters running in SCSI mode. The series of tests included database TPC-C-like tests using Microsoft SQL Server 2019 and Oracle Database 19c running on Red Hat Enterprise Linux 8 (RHEL 8). The Emulex LPe35002 HBA running in NVMe/FC mode demonstrated superior performance in both test scenarios.
Storage
|
Broadcom Inc.
Emulex® LPe35002 Host Bus Adapter
32G Fibre Channel HBA Performance vs. Marvell QLogic QLE2772
Document number:
220114
Release Date:
24 Apr 2020
Innovations in server and storage technologies such as all-flash arrays, NVMe, and NVMe over Fabrics (NVMe-oF) are delivering record speeds and lower latency to improve application performance. As a result, enterprise datacenters require high performance Gen 7 32G Fibre Channel (32GFC) throughput for server connectivity that also provides extreme reliability and scalability under load. Additionally Gen 7 Fibre Channel maintains backward compatibility with existing Fibre Channel infrastructure to ensure streamlined performance upgrades.
Broadcom commissioned Tolly to benchmark the performance of its Emulex Gen 7 LPe35002 32GFC host bus adapter (HBA) against the performance of Marvell’s QLogic QLE2772 32GFC HBA (branded as “enhanced 32GFC”), running database applications and synthetic benchmarks. Both products are second generation, dual- port, 32GFC HBAs. The series of enterprise datacenter-oriented tests performed included synthetic IO, latency, bandwidth, and database TPC-C-like tests.
The Emulex LPe35002 HBA demonstrated superior performance across all test scenarios including up to 5x higher IOPS, as much as 1/2 the latency, and approximately 20% more transactions per minute (TPM) for Microsoft SQL Server and Oracle Database OLTP workloads.
Storage
|
Dell Technologies
Webinar - Future Proof Your VMware Cloud Foundation Infrastructure for Uncertain Times
Document number:
220119
Release Date:
16 Apr 2020
Requirements for change on an IT Infrastructure are at an all time high. Unprecedented demand for work from home means further strain on often already strapped server infrastructure.
In this session we actively discuss how customers have been coping with these critical business demands and how to, not only deliver mission critical applications with changing requirements on IT, but also plan for future changes.
We will also discuss how VMWare Cloud Foundation 4.0 validated design on PowerEdge MX delivers cloud infrastructure that is built for unique innovation.
View the on-demand webcast with Kevin Tolly, Founder at The Tolly Group presenting the study results along with Seamus Jones, Technical Marketing and Roger Foreman, Software Product Marketing at Dell Technologies by clicking on the link below which will take you to the Dell webcast landing page.
Dell Webcast View On-demand Landing Page.
The Tolly report can be downloaded from
here.
Storage
|
Dell Technologies
Tolly Proven PowerEdge MX Reference Architecture for VMware Cloud Foundation
and Deployment Best Practices
Document number:
220102
Release Date:
06 Apr 2020
Large data centers are by nature complex. Deploying and managing data centers consisting of hundreds of compute and storage resources and likely thousands of virtual machines (VM) is no simple task, however. What you need for hybrid cloud is a single integrated solution that is easy to deploy and operate through built-in automated lifecycle management. What you need is VMware Cloud Foundation.
VMware Cloud Foundation is the capstone to the VMware product suite that delivers the data center of today and of the future. Cloud Foundation was born to run on Dell Technologies PowerEdge MX7000 modular, software-defined server infrastructure which, importantly, is a VMware Validated Design (VVD). The PowerEdge MX7000 is a chassis with its eight slots for compute and/or storage resources (sleds) meshes perfectly with VMware’s recommended cluster size of four hosts. Our intent is to describe in a realistic fashion how much infrastructure (hardware), time, and capital investment is required to get it off the ground and expand it down the road. VVD is a set of stringent design criteria established by VMware for SDDC.
In this project, we will build out a production-ready Cloud Foundation built on the Dell PowerEdge MX Modular Server Infrastructure. Our initial system will be built in a single chassis and provide a fully functional management cluster of four hosts running VMware SDDC Manager (Software-defined data center) along with all related VMware features and functions needed to deploy and manage a sophisticated hybrid cloud data center.
Then, we will outline the steps and changes required to grow from our single chassis to a multi-chassis deployment where two or more chassis build upon the initial Cloud Foundation installation. In fact, this solution is easily expandable to up to 10 PowerEdge MX chassis within the same low latency, inter-chassis, scalable network fabric.
The Tolly Group worked closely with Dell Technologies to document a “Tolly Proven” reference architecture/solution for leveraging Dell PowerEdge MX for private cloud deployment. The conceptual guidance provided in this document works hand- in-hand with Dell’s step-by-step cookbook on the topic: “Dell EMC VMware Cloud Foundation for PowerEdge MX700 - Deployment Guide” (Cloud Foundation Cookbook for short).
Storage
|
Dell Technologies
Webinar - Dell iDRAC9 v4.0 Telemetry Streaming
Document number:
220109
Release Date:
21 Feb 2020
The datacenter of the future will rely on artificial intelligence (AI) to drive predictive & proactive management of the server infrastructure. In turn, AI systems rely on ingesting massive amounts of granular data. iDRAC9’s Telemetry Streaming, available with the new iDRAC9 Datacenter license, is architected to provide highly granular, high-value data using resource-efficient data streaming.
View the on-demand webcast with Kevin Tolly, Founder at The Tolly Group presenting the study results along with Rick Hall, Systems Management Product Planning and Strategy, Doug Iler, iDRAC Product Manager and Michael Brown, Distinguished Engineer at Dell Technologies by clicking on the link below which will take you to the Dell webcast landing page.
Dell iDRAC9 v4.0 Webcast View On-demand Landing Page.
The Tolly report can be downloaded from
here.
(Note: There is no download for this webcast item.)
Storage
|
Dell Technologies
iDRAC9 Telemetry Streaming
Evaluation of The Performance and Efficiency of Telemetry Streaming in the New iDRAC9 v4.0 Release
Document number:
220101
Release Date:
18 Feb 2020
The datacenter of the future will rely on artificial intelligence (AI) to drive
predictive & proactive management of the server infrastructure. In turn, AI
systems rely on ingesting massive amounts of granular data. iDRAC9’s
Telemetry Streaming, available with the new iDRAC9 Datacenter license, is
architected to provide highly granular, high-value data using resource-efficient
data streaming.
Dell EMC commissioned Tolly to benchmark the number of data points
collected from a representative PowerEdge server via iDRAC9 v4.00.00.00 in a
24-hour period and compare that to a prior, less data-rich monitoring
implementation. Additionally, data collection via polling and streaming were
compared for network efficiency.
Tolly found that iDRAC9 Telemetry Streaming running on a representative test
system produced over 2.9M data points in a 24-hour period compared with
6,240 data points for a typical polled server using the existing Redfish interfaces
and a reasonable real-world polling interval of 30 minutes. With Telemetry
Streaming each 24-hour report can be collected via a single HTTP request
compared with a per-report best case of 17,280 HTTP requests to collect a
single report daily.
Storage
|
Mellanox Technologies, LTD
Mellanox ConnectX®-5 25GbE Ethernet Adapter Performance vs Broadcom NetXtreme® E
Document number:
219130
Release Date:
07 Oct 2019
Data center computing environments for Cloud, Edge Computing and Hyperconverged
infrastructures (HCI) demand high performance throughput, 25GbE or
higher server connectivity, scalability and resilience.
Mellanox commissioned Tolly to benchmark the performance of its ConnectX-5
25GbE network adapter against the performance of Broadcom’s NetXtreme E
25GbE network adapter in cloud and storage computing environments. The series
of datacenter-oriented tests used included RoCE fabric, FIO, DPDK RFC 2544 and
DPDK CPU throughput.
The Mellanox ConnectX-5 25GbE adapter demonstrated higher throughput, better
scale and lower resource utilization across the set of test scenarios.
Storage
|
Broadcom Inc.
Broadcom NetXtreme® 25GbE Adapter
Ethernet Adapter Impact on vSAN Performance: Competitive Comparison
Document number:
219123
Release Date:
04 Jun 2019
Hyperconverged infrastructure (HCI) combines virtualization, computer servers,
the storage network and storage to provide a simplified architecture that offers
both agility and lower total cost of ownership (TCO). HCI implementations such
as VMware vSphere/vSAN can be optimized and deliver higher performance
using 25GbE networking.
Broadcom commissioned Tolly to benchmark the vSAN storage networking
performance of the Broadcom NetXtreme 25Gb Ethernet Adapter in a VMware
hyperconverged environment and compare those results to similar Ethernet
adapters from Intel, Marvell and Mellanox.
The Broadcom NetXtreme adapter outperformed the competing solutions in
both IOPS and data throughput in all scenarios of 16KB blocks and higher.
Storage
|
Hewlett Packard Enterprise
Proof of Concept - HPE Synergy Composable Infrastructure
Document number:
217120
Release Date:
23 Aug 2017
All hyperbole aside, “Composable Infrastructure”
offers an approach to server provisioning that is a
radical departure from prior solutions - and with
radical benefits. HPE Synergy helps customers reduce
complexity, deploy more rapidly and bring cloud-like
speed to the corporate datacenter.
HPE commissioned Tolly to run common Proof of
Concept (POC) test scenarios that would be applicable
to most infrastructure customers. Tolly engineers worked through
nearly 20 different scenarios across four key areas: infrastructure,
stateless boot/streaming, storage provisioning, and network fabric
configuration. The goal in each of these scenarios was to validate
important capabilities of HPE Synergy.
Storage
|
Load DynamiX, Inc.
Podcast: Flash Storage Benchmarking with Load DynamiX
Document number:
215143
Release Date:
13 Jul 2015
Flash Storage is increasingly important in Enterprise and cloud computing, while still relatively expensive, it can provide significant performance benefits. To be certain that the flash-based array is delivering up to its specifications, one needs to benchmark it.
Load DynamiX is a leader in the storage benchmarking space and in this podcast Tolly Group Founder Kevin Tolly discusses this important topic with Len Rosenthal, VP of Marketing, Load DynamiX.
Storage
|
Hewlett Packard Enterprise
HP StoreEasy Storage:
System Deployment, Ease-of-Use and Management Evaluation
Document number:
214129
Release Date:
15 Sep 2014
From the smallest branch office to the largest enterprise environment, storage is always an important consideration. Nowadays with audio files and video files becoming commonplace elements of business, storage demands reach terabyte levels even with a small number of users. HP believes that, across the board, businesses need efficient file storage that is both easy-to-manage and scalable.
HP commissioned Tolly to evaluate the HP StoreEasy Storage family of devices which
consists of the HP StoreEasy 1000 Storage file appliances to provide simple file storage
and the HP StoreEasy 3000 Gateway Storage file gateways to provide scalable
connectivity to back-end storage area network (SAN) environments. Specifically, Tolly
focused on initial deployment, ease-of-use and management of both systems.
Storage
|
Fusion-io, Inc.
Webcast: Fusion-io Hybrid Storage -
The Proof is in the Performance
Document number:
214109
Release Date:
01 Feb 2014
LIVE DATE: January 23,2014
Speaker:
John Tolly, Director of Engineering, Tolly
Webcast Summary:
Listen to John Tolly from Tolly Group as he discusses their independent testing results comparing the performance of two different hybrid storage systems – Fusion ioControl Hybrid Storage using PCIe flash and a conventional hybrid architecture using NVRAM plus SSDs. You’ll learn how and where the performance of these two hybrid architectures differ and what it means to your application environment.
Hear John describe the test environment and the results, including comparisons of VM density, handling VDI boot storms, IOPS, latency, and various read/write workloads. Register now to understand the type of results you can achieve with hybrid storage systems. The webinar will feature live Q&A after the presentation.
View the on-demand Webcast by clicking the link below:
View Webcast
Storage
|
Fusion-io, Inc.
Fusion ioControl n5 Hybrid Storage Architecture
Performance Comparison versus Conventional NVRAM+SSD+Disk Architecture
Document number:
213154
Release Date:
15 Oct 2013
Conventional hybrid storage solutions use NVRAM (non-volatile RAM/flash) to cache data before writing to disk, then augment performance with an SSD read cache. While such solutions generally offer acceptable performance for small datasets, they are ultimately limited by the configuration and number of SSDs
that can be deployed. In contrast, a hybrid storage approach can leverage PCIe based flash to provide better performance. This not only provides direct (non-SATA/SAS) access to block-level storage and fast read/write caching, but also allows for greater disk capacity per array shelf.
Fusion-io commissioned Tolly to evaluate the performance of its ioControl Hybrid storage approach against a similarly-equipped conventional NVRAM+SSD+Disk hybrid solution. Tolly found the Fusion-io solution to offer high performance and scalability. In block level access tests, Fusion-io provided up to 7X better performance than the conventional NVRAM+SSD+Disk solution
under test.
Storage
|
Overland Storage
Overland Storage SnapScale X2 Network Attached Storage Performance and Scalability Evaluation
Document number:
213140
Release Date:
27 Sep 2013
Traditional scale-up Network Attached Storage (NAS) architecture uses a single head unit to control and protect a number of disk arrays. Although popular, this approach introduces a single point of failure and potentially creates bottlenecks as additional capacity is added. Scale-out architectures, on the other hand, leverage multiple head units (nodes) that work together to protect data, and can expand easily to spread processing power and bandwidth to maximize performance and redundancy.
Overland Storage commissioned Tolly to evaluate the scalability and performance of their new SnapScale X2 clustered NAS solution. Tolly evaluated the SnapScale X2 platform in sequential and random read/write test scenarios. Tolly validated that the SnapScale X2 can scale to 50 nodes seamlessly, with no disruption. Adding nodes increased the capacity and also improved cluster-wide performance as additional nodes were added to the cluster.
Storage
|
Quantum Corporation
Quantum vmPRO Data Protection Solution for VMware
An Evaluation of Ease-of-Implementation, Functionality, and Interoperability in VMware vSphere 5 Environments
Document number:
212100
Release Date:
02 Feb 2012
With virtualization’s ever-increasing adoption by small and medium size businesses (SMBs), and remote or branch offices (ROBOs), IT administrators face a major overhaul of the critical infrastructure that supports the expanding virtual environment.
In the data protection arena, while the environment will shift, the goal remains the same: to provide backup and restore capabilities to every server instance both as a whole, and as a file structure for individual recovery. The task then becomes choosing and implementing data protection for virtualized environments while maintaining interoperability and manageability with existing disaster recovery deployments.
Storage
|
Hewlett Packard Enterprise
HP X5000 G2 Network Storage System
System Ease-of-Use, High-Availability, Performance, and Interoperability Evaluation
Document number:
211130
Release Date:
06 Dec 2011
Today’s small and medium-size businesses (SMBs) have ever-growing needs with respect to storage. Not only do SMBs often need many terabytes of storage, they require a level of high-availability (HA) typically associated with large-scale, enterprise deployments. Furthermore, given IT staffing constraints, storage solutions should be easily installed and managed by existing IT staff.
Tests show that a 2-node, HP X5520 G2 9.6TB LFF 15K Network Storage System (hereafter X5000 G2) could be installed “out-of-the-box” to a fully-functional, HA system joined to a Microsoft domain in less than 90 minutes. The HP-supplied Initial Configuration Tasks tool guides the user through each deployment stage, eliminating a great deal of manual configuration.
Storage
|
Quest Software, Inc.
Quest vRanger Pro 5.2
Performance and Ease-of-Use Evaluation in VMware ESXi 4.1u1 Environments
Document number:
211134
Release Date:
05 Dec 2011
Small and medium size business (SMB) networks are facing increasing demands for data protection in their virtualized environments. As server consolidation continues, SMBs are unlikely to adhere to the level of redundancy sought by enterprise class datacenters. However, planning for the unexpected provides for a quick recovery in the event of an outage.
All SMBs can benefit from a flexible platform which can provide the desired backup, recovery, and replication features, while effortlessly integrating with their specific environment. A solution which can be implemented as a virtual or physical machine, on the LAN or SAN is preferable where storage may not be centralized, or additional servers may be hard to come by.
Storage
|
Hewlett Packard Enterprise
HP StorageWorks X3820 2-node Network Storage System:
High-Availability Storage for SMB
System Evaluation
Document number:
210122
Release Date:
10 Jun 2010
Today’s small and medium-size businesses (SMBs) have ever-growing needs with respect
to storage. Not only do SMBs need to be able to scale to many terabytes of storage, they
require a level of high-availability typically associated with large-scale, enterprise
deployments. Furthermore, given IT staffing constraints, they need such storage
solutions to be able to be installed quickly by existing IT staff.
HP commissioned Tolly to evaluate the ease-of-installation, high-availability, performance and system support of the HP StorageWorks X3820 2-node Netwok Storage System in the context of SMB deployment.
Storage
|
|
|
Hitachi Data Systems
HDS High-performance NAS Platform 3200 Series with 10GbE, powered by BlueArc
Throughput Evaluation Using the Common Internet File System ( CIFS) Protocol
Document number:
208341
Release Date:
09 Mar 2009
Hitachi Data Systems commissioned The Tolly Group to examine the I/O performance characteristics of its High-performance NAS Platform 3200 Series (HNAS 3200). The HNAS 3200 is a purpose-built network-attached storage (NAS) solution that incorporates the BlueArc Titan 3210 Network Storage System under a strategic partnership between Hitachi Data Systems and BlueArc.
Tolly Group engineers measured the throughput of a single HNAS 3200 node when handling file-access requests using Common Internet File System (CIFS) protocol. The back-end storage system used for testing was Hitachi Data Systems (HDS) Universal Storage Platform V (USP V) outfitted with 256 Fiber Channel HDDs. Throughput was measured for a variety of read, write and mixed read/write operations from/to back-end disks with sequential access. Engineers also measured the throughput of two HNAS 3200 nodes in a cluster in the same scenarios.
Storage
|
Hitachi Data Systems
HDS High-performance NAS Platform 3200 Series with 10GbE, powered by BlueArc
Throughput Evaluation Using Network File System (NFS) Ver. 3 Protocol
Document number:
208340
Release Date:
09 Mar 2009
Hitachi Data Systems commissioned The Tolly Group to examine the I/O performance characteristics of its High-performance NAS Platform 3200 Series (HNAS 3200). The HNAS 3200 is a purpose-built network-attached storage (NAS) solution that incorporates the BlueArc Titan 3210 Network Storage System under a strategic partnership between Hitachi Data Systems and BlueArc.
Tolly Group engineers measured the throughput of a single HNAS 3200 node when handling file-access requests using Network File System (NFS) Version 3 protocol. The back-end storage system used for testing was Hitachi Data Systems (HDS) Universal Storage Platform V (USP V) outfitted with 256 Fiber Channel HDDs. Throughput was measured for a variety of read, write and mixed read/write operations from/to back-end disks with sequential access. Engineers also measured the throughput of two HNAS 3200 nodes in a cluster in the same scenarios.
Download the free report.
Storage
|
BlueArc
BlueArc Titan 3210 Network Storage System Throughput Evaluation with the Common Internet File System (CIFS) Protocol
Document number:
208352
Release Date:
06 Mar 2009
BlueArc Corporation commissioned The Tolly Group to examine the I/O performance characteristics of the Titan 3210 Network Storage System, a purpose-built network storage solution.
Tolly Group engineers measured the throughput of a single Titan 3210 node when handling file-access requests using the Common Internet File Systems (CIFS). The back-end storage system used for testing was Hitachi Data Systems (HDS) Universal Storage Platform V (USP V) outfitted with 256 Fibre Channel (FC) HDDs. Throughput was measured for a variety of read, write and mixed read/write operations from/to back-end disks with sequential access. Engineers also measured the throughput of two Titan 3210 nodes in cluster in the same scenarios.
BlueArc’s Titan 3210 Network Storage System delivered near wire-speed, 10-GbE throughput conducting sequential read operations for various block sizes using CIFS file system and doubled its performance in the same tests by clustering two nodes.
Tests were conducted in October 2008.
Download the free report.
Storage
|
BlueArc
BlueArc Titan 3210 Network Storage System Throughput Evaluation with NFS Version 3 Protocol
Document number:
208351
Release Date:
20 Nov 2008
BlueArc Corporation commissioned The Tolly Group to examine the I/O performance characteristics of the Titan 3210 Network Storage System, a purpose-built network storage solution.
Tolly Group engineers measured the throughput of a single Titan 3210 node when handling file-access requests using the Network File Systems (NFS) Version 3 protocol. The back-end storage system used for testing was Hitachi Data Systems (HDS) Universal Storage Platform V (USP V) outfitted with 256 Fibre Channel (FC) HDDs. Throughput was measured for a variety of read, write and mixed read/write operations from/to back-end disks with sequential access. Engineers also measured the throughput of two Titan 3210 nodes in cluster in the same scenarios.
BlueArc’s Titan 3210 Network Storage System delivered near wire-speed, 10-GbE throughput conducting sequential read operations for various block sizes using NFS file system and doubled its performance in the same tests by clustering two nodes.
Tests were conducted in October 2008.
Download the free report.
Storage
|
Hewlett Packard Enterprise
Hewlett-Packard Co.
Hardware-Accelerated iSCSI Solutions
Performance Comparison of Hardware and Software Accelerated iSCSI from ProLiant Servers to StorageWorks Storage Array
Document number:
208305
Release Date:
16 Jul 2008
Hewlett-Packard Co. commissioned The Tolly Group to evaluate the performance benefits of multifunction server adapters in ProLiant servers featuring hardware-accelerated iSCSI over traditional software-based iSCSI initiators communicating to StorageWorks storage arrays.
Tolly Group engineers tested two end-to-end iSCSI storage solutions from HP. The first was a ProLiant server with an HP NC373T Gigabit Server Adapter communicating with a StorageWorks 1200 All in One (AiO) Storage System. Engineers also tested a ProLiant DL380 G5 server with an
HP NC373i Gigabit Server Adapter communicating with a StorageWorks 2012i Modular Smart Array (MSA2012i). Tests compared the throughput and CPU utilization of software-based iSCSI (Microsoft iSCSI initiator) against hardware-accelerated iSCSI under various disk access scenarios, using and Iometer test tool.
Tests were conducted in April 2008.
Storage
|
Hewlett Packard Enterprise
Hewlett-Packard Co.
StorageWorks 8Gb Simple SAN Connection Kit
Comparison of Performance and Ease of SAN Deployment using HP Simple SAN Connection Manager (SSCM) Solution
Document number:
208276
Release Date:
20 Jun 2008
Hewlett-Packard Co. commissioned The Tolly Group to evaluate the performance of its Fibre Channel-based Storage-Works 8Gb Simple SAN Connection Kit compared to a traditional 4Gb Fibre Channel (FC) solution.
The Tolly Group certified HP’s 8Gb Fibre Channel SAN solution featuring the StorageWorks 8Gb Simple SAN Connection Kit to be the “First 8Gb Fibre Channel SAN Solution” tested. Tolly Group engineers also compared the ease of deployment of a singlevendor end-to-end FC SAN, using HP’s Simple SAN Connection Manager, to that of a traditional multivendor FC SAN that required individual components like the HBA, switch and storage subsystem to be configured individually. Tests focused on the number of configuration steps and variables that needed to be configured to finish the deployment of the SAN.
Tests were conducted in February 2008.
Storage
|
Linksys, Inc.
Network Storage System NSS6100
Four Bay, 1TB Advanced Gigabit Chassis with RAID
Competitive Performance Evaluation of Network-
Attached Storage Systems
Document number:
208254
Release Date:
20 Feb 2008
Linksys, a division of Cisco Systems, Inc., commissioned The Tolly Group to evaluate the performance and features of the NSS6100 network-attached storage appliance against NETGEAR’s ReadyNAS™ 1100, and ReadyNAS™ NV+; along with Buffalo Technology (USA), Inc.’s TeraStation PRO™ II appliances.
The Tolly Group examined the NAS appliances’ multi-client file server performance using Ziff Davis Media’s Net-Bench 7.0.3 benchmarking test suite. As a baseline to compare against other
single-client I/O tests, engineers also evaluated the single client read-and-write mode I/O performance using IOZone benchmark version 3.282. Tests focused on RAID 1, RAID 5 and JBOD configurations, and with on-disk encryption enabled, when supported. Tests were conducted in December 2007.
Storage
|
|
|
QLogic Corp.
"Talking Outside the Box": Podcast Interview with QLogic's Tim Lustig and Overview of Tolly Up-to-spec Testing
Document number:
207203
Release Date:
01 Feb 2007
This 14-minute podcast focuses on Fibre Channel Storage Area Networking (SAN). Tolly Group Pres/CEO, Kevin Tolly, provides an overview of a recent Tolly Group Up-to-spec test of QLogic’s SANbox5600.
In addition, Charlie Bruno, Executive Editor of The Tolly Group, discusses the ramifications of the test with QLogic's Tim Lustig. Details of the test can be found in Tolly Group document 206124.
Storage
|
|
|
QLogic Corp.
Tolly Group White Paper Series– Architecture in Action: Building SANs with Stacks
Document number:
206124
Release Date:
08 Aug 2006
QLogic commissioned The Tolly Group to build a microcosm of an SMB/Enterprise Fibre Channel environment to quantify critical elements related to scalability and management of competing SAN approaches, as well as to analyze the economic impact of each. This white paper provides a hands-on assessment of the scalability issues and the cost-of-ownership for the QLogic SANbox 5000 versus rival products that utilize a different architectural approach to SANs.
QLogic SANbox 5000 series “stackable” switches provide high-speed 10-Gigabit per second (Gbps) links dedicated to linking individual physical switches without reducing the number of ports available for use by end-user devices. In effect, this approach creates a virtual backplane bringing the performance advantages of the chassis approach to low-cost fixed-port switches. The white paper is based on two hands-on examinations conducted by The Tolly Group.
In the first part, The Tolly Group “built out” an actual SAN using two alternative offerings. The goal was to illustrate the configuration effort and to quantify hardware/software requirements associated with growing a SAN from an “entry” level connecting just two Fibre Channel devices through to a point where scores of devices could be connected.
In the second part of the study, engineer/analysts built and validated a cost model comparing the QLogic solution versus several leading “non-stackable” alternatives.
Storage
|
BlueArc
BlueArc Corp. Titan 2200 Network Attached Storage System, Throughput Evaluation Under Mixed I/O Scenarios
Document number:
206132
Release Date:
15 May 2006
BlueArc Corp. commissioned The Tolly Group to examine the performance of the company’s Titan 2200, a network-attached storage (NAS) solution that the company claims is the fastest system on the market and also the most scalable, able to consolidate and manage up to 512 terabytes of data in a single storage pool.
Tolly Group engineers focused tests on the average throughput of a single Titan 2200 node handling I/O requests between 48 client PCs. Throughput was measured for a variety of read, write and mixed read/write operations.
Storage
|
Tacit Networks, Inc. (Packeteer/Blue Coat)
Tacit Networks Ishared Server and Ishared Remote Appliances, Performance and Functionality Evaluation
Document number:
205133
Release Date:
25 Oct 2005
Tacit Networks, 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 “Stackable” Services. Tacit Networks’ Ishared product line consists of network appliances designed to deliver wide area file services (WAFS) and other branch office IT services on a single consolidated platform.
Tolly Group engineers conducted extensive performance tests designed to measure the effectiveness of the Tacit Networks appliances – in terms of the response time improvements delivered to users and the yield in bandwidth reduction from caching frequently requested files and E-mail with attachments on local Ishared Remote devices. 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 Tacit Networks facilities in South Plainfield, NJ.
Storage
|
Maranti Networks, Inc. (EMC)
Maranti, Inc. CoreSTOR 2000 Network Storage Controller SAN Performance
Document number:
204143
Release Date:
19 Oct 2004
Maranti, Inc. commissioned The Tolly Group to evaluate its CoreSTOR 2000™ Network Storage Controller, a modular, high-performance network storage controller that delivers 8 to 16 ports across a 3U chassis. CoreSTOR 2000 delivers wire-speed storage virtualization and native CoreSTOR storage services to heterogeneous storage arrays and servers across all ports on the chassis. By supporting Gigabit Ethernet and Fibre Channel ports across the chassis, CoreSTOR 2000 provides connectivity between existing Fibre Channel and emerging iSCSI environments.
Tolly Group engineers measured the performance of the CoreSTOR 2000 in a simulated Exchange and SQL server environment and in three main areas: the I/O processing rate (I/O per second or IOPS), remote mirroring performance and high availability of the storage network controller when subjected to a number of scenarios.
Tests show that the Maranti CoreSTOR 2000 delivers consistent I/O processing performance; in fact, IOPS performance actually increases as the number of servers was ramped up from one, to four and then eight. Moreover, the CoreSTOR 2000 increased performance in most cases, as the number of local and remote mirrored storage subsystems increased. Interestingly, as loading increased on the CoreSTOR 2000, response times climbed as the number of servers increased, due to more I/O requests to the controller of the storage arrays. Response times decreased, however, as the number of storage subsystems increased because the requests were spread out to all available storage arrays via Maranti proprietary load-balancing scheme. Tests were conducted during August 2004.
Storage
|
User not logged in