Dell PowerEdge C5220 - Principled Technologies

DELL POWEREDGE C5220: HADOOP PERFORMANCE
Every year, the amount of data that businesses must sort through grows
enormously. The ability to sort, filter, and analyze this data is becoming more
and more vital to many businesses in analyzing their customers and their market
segment. Additionally, businesses need an infrastructure that is powerful and
flexible, but also compact and scale-friendly. The Dell PowerEdge C5220 server is
an ideal solution to pair with Apache Hadoop, a powerful multi-node data
analysis application. With the PowerEdge C5220, organizations can scale out to
their data processing requirements and successfully handle these ever-increasing
data volumes, finding new value in their big data.
To test the Hadoop performance capabilities of the Dell PowerEdge
C5220, we configured eight Dell PowerEdge C5220 servers into a Hadoop cluster
and ran the TeraSort benchmark on the platform. We found that eight Dell
PowerEdge C5220 servers, all contained within the single shared infrastructure
design of the Dell PowerEdge C5000 chassis, could sort a 10GB dataset in just
155 seconds.
A PRINCIPLED TECHNOLOGIES TEST REPORT
Commissioned by Dell Inc.; April 2012
DELL POWEREDGE C5220 AND HADOOP – BETTER TOGETHER
Today’s marketplace has moved beyond the structural confines of relational
data and up to the category of “big data.” Companies with hyperscale data centers are
dealing with tremendous amounts of existing data, with massive data growth on the
horizon. They require reliable and powerful hardware and software platforms on which
to process this big data. Dell PowerEdge C series servers provide this solid infrastructure
for companies to offer their data processing capabilities. Integrating many PowerEdge C
servers together in a Hadoop cluster allows companies to process large datasets quickly
and efficiently.
In our lab tests, we configured eight Dell PowerEdge C5220 microservers into a
Hadoop cluster and ran the TeraSort benchmark on the cluster to demonstrate the
performance capabilities of the Dell PowerEdge C platform. Below, we discuss the
PowerEdge C5220 microserver, our test topology, and our results in more detail.
Figure 1: The Dell PowerEdge
C5220 microserver provides
power and reliability in a tight
and dense form factor, within
the C5000 shared
infrastructure.
Dell PowerEdge C5220
The Dell PowerEdge C5220
microserver is a highly dense small-formfactor server mounted on a sled, which
is housed in a Dell PowerEdge C5000
shared infrastructure enclosure, offering
up to 12 microservers in a 3U rack
density. These high physical server
densities offer massive scale-out
configurations, and are ideal for service
providers and large data centers that
need to optimize their space, power, and
Highlights of the Dell
PowerEdge C5220
 Up to 12 PowerEdge C5220 microservers
in a single 3U C5000 chassis
 Cold aisle serviceability
 Shared infrastructure uses less floor
space, power, and cooling
 Dense form factor ideal for service
providers, hosting platforms, and
hyperscale environments
 Up to 6TB raw storage in a single
microserver
 Embedded chassis management control
cooling infrastructure. Each server, all coldaisle accessible for maintenance, features processing power from the Intel® Xeon®E31200 series , up to 32GB RAM capacity, and up to four drives per server. The high
Dell PowerEdge C5220: Hadoop performance
A Principled Technologies test report 2
performance levels it delivers, coupled with a simple, modular, and easy-to-maintain
design, make the Dell PowerEdge C5220 microserver an optimal choice to deploy for
extremely dense compute fabrics handling big data, large software as a service (SaaS)
environments, and cloud deployments. Figure 2 presents a view of the Dell PowerEdge
C5220.
Figure 2: The PowerEdge C5220 microserver.
The Dell PowerEdge C5220 can deliver many benefits to your large cloud
deployment, and lets you:

Maintain your environment easily with cold-aisle accessible, swappable
microserver modules and connections

Make better use of expensive data center space by increasing the rack
density with up to 12 server nodes fitting into a standard 3U rack slot

Utilize up to 6 TB of raw disk storage per microserver with SAS drives or 4 TB
of raw disk storage per microserver with SATA drives – up to 72 TB and 48
Dell PowerEdge C5220: Hadoop performance
A Principled Technologies test report 3
TB respectively across the entire C5000 shared infrastructure in just a 3U
form factor

Reuse or repurpose servers easily when workloads change with hot-swap
server nodes – you no longer need to experience downtime by replacing the
entire server chassis.
Designed with power efficiency and maintainability in mind, the Dell PowerEdge
C5220 maximizes operating efficiency with a shared-infrastructure design. To learn
more about the Dell PowerEdge C5220 and the entire Dell PowerEdge C Series, visit
http://www.dell.com/us/enterprise/p/poweredge-cloud-servers.
WHAT WE TESTED
To test the ability of the PowerEdge C5220 microserver to handle large data
processing tasks, we used Hadoop, specifically Cloudera Distribution Including Apache
Hadoop (CDH). Below we briefly discuss Hadoop and the benchmark tool we used,
TeraSort.
Hadoop
Hadoop, developed by Apache Software Foundation, is an open-source
distributed application that enables the analysis of large volumes of data for specific
purposes. Using Hadoop’s framework, IT organizations and researchers can build
applications that tailor the data analysis to specific needs for each company, even using
unstructured data. Many different markets—among them finance, IT, and retail—use
Hadoop due to its ability to handle heterogeneous data, both structured and
unstructured.
Hadoop can run across any number of machines using varied hardware,
spreading data across all available hardware resources using a distributed file system,
Hadoop Distributed File System (HDFS), and replicating data to minimize loss if a
hardware malfunction occurs. The software is able to detect hardware failures, and to
work around said failures to allow uninterrupted access to data. Because of its ability to
run on different hardware, a Hadoop cluster is scalable and flexible – it can be expanded
to encompass growing databases and companies. It is also cost-effective, as it allows
companies to utilize commodity hardware effectively.
TeraSort
The process of sifting and sorting through large amounts of data is a critical one
for many businesses, and they need the most efficient set of hardware and software
tools to do the job. The TeraSort benchmark on Hadoop tests the sorting speed and
efficiency of a Hadoop cluster. It measures how quickly a set of systems, in our case
eight PowerEdge C5220 servers, can sort a set amount of data. The main output of a
Dell PowerEdge C5220: Hadoop performance
A Principled Technologies test report 4
TeraSort benchmark run is the time it takes to sort that amount of data. We report
those times, and the specifics of our test scenario, in the sections below.
TEST RESULTS IN MORE DETAIL
We used the TeraSort benchmark program from the Cloudera Hadoop
distribution to generate 10GB of test data and to sort it using an eight-node Hadoop
cluster consisting of Dell PowerEdge C5220 servers. We set up an auxiliary server for the
Cloudera Manager and used the Dell PowerEdge C5220 servers to perform the sort. For
more details on the setup of the Hadoop cluster, see Appendix B.
There are three main portions of the TeraSort benchmark: TeraGen, TeraSort
and TeraValidate. TeraGen generates the data with which to populate your Hadoop
cluster. TeraSort measures the data sort time. When the test run is finished,
TeraValidate validates the sorted output, and reports the time it took to sort the data.
Comparing these times to those of other Hadoop clusters indicates how your
hardware’s sort capability measures up. Figure 3 presents the time, in seconds, it took
for the Dell PowerEdge C5220 cluster to complete each TeraSort task.
TeraSort phase
Generating the data
Sorting the data
Validating the data
Time to complete (seconds)
82
155
59
Figure 3: The time spent in each phase of the Hadoop TeraSort benchmark.
CONCLUSION
Companies that wish to process big data in scale-out environments require
powerful and dense hardware that complete large data processing tasks, such as
Hadoop analysis, quickly and efficiently. Selecting the right server for your underlying
hardware infrastructure is critical at hyperscale.
In our tests, an eight-node Dell PowerEdge C5220 Hadoop cluster was able to
sort the 10 GB of data in just 155 seconds. While each scenario and each environment
vary, our eight-node Hadoop TeraSort test shows that even a small number of Dell
PowerEdge C5220 microservers can sort large amounts of data; as with most Hadoop
operations, we expect adding server nodes would result in an even faster sort time.
Dell PowerEdge C5220: Hadoop performance
A Principled Technologies test report 5
APPENDIX A – SERVER CONFIGURATION INFORMATION
Figure 4 provides detailed configuration information for the microserver we tested.
System
General
Number of processor packages
Number of cores per processor
Number of hardware threads per core
System power management policy
CPU
Vendor
Name
Model number
Stepping
Socket type
Core frequency (GHz)
Bus frequency
L1 cache
L2 cache
L3 cache (MB)
Platform
Vendor and model number
Motherboard model number
BIOS name and version
BIOS settings
Memory module(s)
Total RAM in system (GB)
Vendor and model number
Type
Speed (MHz)
Speed running in the system (MHz)
Timing/Latency (tCL-tRCD-tRP-tRASmin)
Size (GB)
Number of RAM module(s)
Chip organization
Rank
Operating system
Name
File system
Kernel
Language
Updates
Graphics
Vendor and model number
Dell PowerEdge C5220: Hadoop performance
Dell PowerEdge C5220
1
4
1
Balanced
Intel
Xeon
E3-1220
D2
LGA1155
3.20
5 GT/s
32 KB + 32 KB (per core)
256 KB (per core)
8
Dell PowerEdge C5220
1KTH4B0
Dell 1.0.11
Defaults
32
Samsung M391B1G73AH0-YH9
PC3L-10600E
1,333
1,333
9-9-9-36
8
4
Double-sided
Dual
CentOS 6.2, x86_64
ext4
2.6.32-220.13.1.el6.x86_64
English
All as of 4/12/2012
AST2050
A Principled Technologies test report 6
System
Graphics memory (MB)
RAID controller
Vendor and model number
Driver version
Cache size (GB)
Hard drive
Vendor and model number
Number of disks in system
Size (GB)
Buffer size (MB)
RPM
Type
Ethernet adapters
Vendor and model number
Type
Driver
USB ports
Number
Type
Dell PowerEdge C5220
128
Intel C204 SATA Raid Controller
N/A
None
Dell ST9500620NS
4
500
64
7.2K
SATA
Intel 82580DB Gigabit Network Connection
Integrated
Intel(R) Gigabit Ethernet Network Driver; igb, 3.0.6-k
1 internal
2.0
Figure 4: Configuration details for the microserver we tested.
Dell PowerEdge C5220: Hadoop performance
A Principled Technologies test report 7
APPENDIX B – HOW WE TESTED
We installed and configured CentOS on an auxiliary server, which provided both basic services and Hadoop
management for the C5220 nodes in the Hadoop cluster. We installed and configured CentOS on the C5220 nodes, using
the C5000’s baseboard management controller (BMC) virtual console feature. We installed the Cloudera Hadoop
Manager on the auxiliary manager server and used it to install Hadoop on the C5220 nodes.
Configuring the auxiliary server
Below, we review the steps necessary to configure the management server, which runs the necessary auxiliary
services and the Cloudera Manager.
1. Install the base-installation of CentOS 6.2 x86_64 following the steps in the section CentOS base installation.
2. Log onto the server as root.
3. Configure one network interface, e.g. eth0, to access a network connected to the Internet with, for example, IP
address 10.1.1.10.
4. Configure one network interface, e.g., eth1, to access the network of C5220 nodes with, for example, IP address
192.168.1.10 and CIDR prefix of 24.
5. Install CentOS updates:
yum update
6. Reboot the server:
shutdown –r now
7. After the system reboots, log onto the server as root.
8. Install Squid as a HTTP proxy for the C5220 nodes:
yum install squid
9. Apply the following patch file, squid.conf.diff, to modify /etc/squid/squid.conf with the patch program:
acl ftp proto FTP
http_access allow ftp
## cluster network
acl intranet src 192.168.1.0/21
http_access allow intranet
10. Install dnsmasq to supply hostname resolution for the nodes:
yum install dnsmasq
11. Edit the configuration /etc/dnsmasq.conf and modify these lines as shown:
interface=eth1
no-dhcp-interface=eth1
12. Add the IP addresses and hostnames of the auxiliary server and C5220 nodes to /etc/hosts on the auxiliary
server. For example:
192.168.1.10
192.168.1.31
192.168.1.32
192.168.1.33
192.168.1.34
192.168.1.35
192.168.1.36
192.168.1.37
192.168.1.38
had01-ctrl01
cl01n01
cl01n02
cl01n03
cl01n04
cl01n05
cl01n06
cl01n07
cl01n08
13. Start the DNS server:
chkconfig dnsmasq on
service dnsmasq start
Dell PowerEdge C5220: Hadoop performance
A Principled Technologies test report 8
14. Install the ntpd time server to provide a common time source for the nodes:
yum install ntpd
15. Modify the configuration file /etc/ntp.conf to use Internet time sources, and to be a time service on the nodes
network. For example:
# Hosts on local network are less restricted.
restrict 192.168.1.0 mask 255.255.255.0
# Use public servers from the pool.ntp.org project.
# Please consider joining the pool (http://www.pool.ntp.org/join.html).
server 0.pool.ntp.org
16. Start the NTP server:
chkconfig ntpd on
service ntpd start
17. Generate an SSH public-key pair with, for example, no password:
ssh-keygen -t dsa -N ""
18. Reboot the server.
Configuring the C5220 Hadoop nodes
1. Install the base-installation of CentOS 6.2 x86_64 following the steps in the section CentOS base installation.
2. Configure one network interface, e.g. em1, to access the Hadoop network of C5220 nodes with, for example, IP
address 192.168.1.31 and CIDR prefix of 24.
3. Configure the system to synchronize its time with the NTP service on the auxiliary server. Replace the server
lines in /etc/ntp.conf with references to the IP address of the auxiliary server:
server 192.168.1.10
4. Configure the yum updater to use the cluster’s HTTP proxy:
echo "proxy=http://192.168.1.10:3128/" >> /etc/yum.conf
5. Copy the auxiliary server’s public SSH key to the node to allow easy remote management:
mkdir /root/.ssh
chmod 700 /root/.ssh
scp 192.168.1.10:/root/.ssh/id_dsa.pub /root/.ssh/authorized_keys
6. Install CentOS updates:
yum update
7. Reboot the server:
shutdown –r now
8. Repeat steps 1 through 7 for the remaining nodes.
Cloudera Hadoop (CDH3u3) installation using Cloudera manager (3.7.5)
1. Log onto the auxiliary server as root.
2. Download and transfer the free edition of the Cloudera Manager installer, version 3.7.5, for CentOS from
Cloudera.com to the auxiliary server.
3. Run the Cloudera Manager installer:
./cloudera-manager-installer.bin
4.
5.
6.
7.
8.
On the Cloudera Manager README screen, read the installation overview, and select Next.
On the Cloudera Manager (Free Edition) License screen, review the license, and select Next.
On the next screen, select Yes to accept this license.
On the Oracle Binary Code License Agreement screen, review the license, and select Next.
On the next screen, select Yes to accept this license.
Dell PowerEdge C5220: Hadoop performance
A Principled Technologies test report 9
9. The auxiliary server now has a Web interface for managing Hadoop. Select OK on the Next Step screen.
10. On the installer’s Finish screen, select OK to exit the installer.
11. Using a Web browser, open the page http://<clouderamanager>:7180/ , where <clouderamanager> is the IP
address of the auxiliary manager server.
12. Copy the SSH keys, previously generated in the Configuring the auxiliary server section, from the auxiliary server
to the local desktop running the Web browser.
13. Log onto the Cloudera manager with username admin and password admin.
14. On the Thank you for choosing Cloudera Manager and Cloudera’s Distribution including Apache Hadoop (CDH)
screen, click Continue.
15. On the register your CDH3 installation, click Skip Registration.
16. On the Specify hosts for your CDH3 cluster installation screen, enter a list of hostnames or IP addresses in the
text box at the bottom, such as 192.168.1.3[1-8]. Click Find Hosts.
17. The upper text box will contain a list of potential Hadoop hosts. Select the applicable nodes, and click Continue.
18. On the provide SSH login credentials screen, select All hosts use the same public key, click Choose file for the
Public Key File, and browse to and select the SSH public-key file for the auxiliary server’s public key (see step 12).
19. Repeat step 18 to select the auxiliary server’s Private Key File.
20. Click Start Installation.
21. After the Hadoop installation on the nodes has finished, switch to a console session on the auxiliary server.
22. Format two of the remaining four disks on each node as part of the Hadoop file system.
# Note: in the following example, the contents of disks /dev/sdc and /dev/sdd
#
will be destroyed
# partition /dev/sdc and /dev/sdd and create EXT4 file systems
for i in $(seq 31 38); do
ssh 192.168.1.$i parted -s /dev/sdc mklab gpt \; parted -s /dev/sdd mklab gpt
ssh 192.168.1.$i parted /dev/sdc mkpart primary '"1 -1"' \; \
parted /dev/sdd mkpart primary '"1 -1"'
ssh 192.168.1.$i mkfs.ext4 /dev/sdc1 \; mkfs.ext4 /dev/sdd1
done
# Modify fstab so that these file systems are mounted at boot time
for i in $(seq 31 38); do
ssh 192.168.1.$i '(echo "/dev/sdc1 /dfs/d1 ext4 defaults,noatime 1 2";\
echo "/dev/sdd1 /dfs/d2 ext4 defaults,noatime 1 2") \
>> /etc/fstab'
done
# Create the default Hadoop HDFS directories
for i in $(seq 31 38); do
ssh 192.168.1.$i mkdir -p '/dfs/{d1,d2,n1,n2,s1,s2}' '/mapred/{local,jt}'
ssh 192.168.1.$i mount '/dfs/d{1,2}'
done
for i in $(seq 31 38); do
ssh 192.168.1.$i chmod 700 '/dfs/{d1,d2,n1,n2,s1,s2}' \;
chmod 755 '/mapred/{local,jt}'
ssh 192.168.1.$i chown hdfs:hadoop '/dfs/{d1,d2,n1,n2,s1,s2}' \;
Dell PowerEdge C5220: Hadoop performance
A Principled Technologies test report 10
chown mapred:hadoop '/mapred/{local,jt}'
done
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
Return to the Cloudera Manager in the browser.
On the Services screen, click Add a Service.
On the Select the type of service you want screen, select HDFS and click Continue.
Accept the default host assignments for HDFS (eight DataNodes and one each of NameNode, Secondary
NameNode and Balancer nodes), and click Continue.
Review the configuration changes, and click Continue.
On the Services screen, click Add a Service.
On the Select the type of service you want screen, select MapReduce and click Continue.
On the Select the set of dependencies for your new service screen, select HDFS and click Continue
Accept the default host assignments for HDFS (eight TaskTrackers and one JobTracker nodes), and click
Continue.
Review the configuration changes, and click Continue.
Running the TeraSort benchmark
1. Log onto one of the Hadoop nodes as root.
2. Switch the login shell to the mapred user:
su - mapred
3. Generate 10GB of data (or 100,000,000 100-byte rows of data) for sorting by the cluster by starting a Hadoop
job. The data will be stored on the HDFS file system under the directory /user/mapred/ts-in.
hadoop jar ~/hadoop-examples.jar
teragen
100000000 /user/mapred/ts-in
4. When that job completes, start the TeraSort job. The sorted data will be found under directory
/user/mapred/ts-out/
hadoop jar ~/hadoop-examples.jar
terasort \
/user/mapred/ts-in /user/mapred/ts-out
5. Finally, verify that the data is properly sorted by running the validation jobs:
hadoop jar ~/hadoop-examples.jar
teravalidate
/user/mapred/ts-in /user/mapred/ts-out
CentOS 6.2 base installation
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
Insert and boot from the CentOS-6.2-x86_64-bin-DVD1 installation DVD.
At the welcome screen, select Install or upgrade an existing system, and press Enter.
At the Media test screen, select Skip, and press Enter.
At the CentOS 6 title screen, click Next.
At the Choose an Installation Language screen, select English, and click Next.
At the Keyboard Type screen, select U.S. English, and click Next.
At the Storage Devices screen, select Basic Storage Devices, and click Next.
If a warning for device initialization appears, select Yes, discard any data.
At the Name the Computer screen, type the host name, and click Configure Network.
At the Network Connections screen, select the server’s main or management network interface, and click Edit.
At the Editing network interface screen, check Connect Automatically.
Dell PowerEdge C5220: Hadoop performance
A Principled Technologies test report 11
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
On the same screen, Select the IPv4 Settings tab, change the Method to Manual, and click Add.
On the same screen, enter the IP address, Netmask, Gateway, and DNS server. Click Apply.
Click Close on the Network Connections screen, and click Next on the Name the Computer screen.
At the Time zone selection screen, select the appropriate time zone, and click Next.
Enter the root password in the Root Password and Confirm fields, and click Next.
At the Partition selection screen, select Replace Existing Linux System(s), and click Next.
If a warning appears, click Write changes to disk.
At the default installation screen, click Next to begin the installation.
At the Congratulations screen, click Reboot.
After the system reboots, log in as root.
Disable SELinux by editing the file /etc/selinux/config, and change the line SELINUX=enforcing to
SELINUX=disabled. These changes take effect after rebooting.
23. Disable these unused services by running the following command-line script:
CHK_OFFs="auditd autofs cups ip6tables iptables nfslock netfs portreserve postfix\
qpidd rhnsd rhsmcertd rpcgssd rpcidmapd rpcbind"
for i in ${CHK_OFFs}; do
chkconfig $i off
service $i stop
done
Dell PowerEdge C5220: Hadoop performance
A Principled Technologies test report 12
ABOUT PRINCIPLED TECHNOLOGIES
We provide industry-leading technology assessment and fact-based marketing
services. We bring to every assignment extensive experience with and expertise
in all aspects of technology testing and analysis, from researching new
technologies, to developing new methodologies, to testing with existing and new
tools.
Principled Technologies, Inc.
1007 Slater Road, Suite 300
Durham, NC, 27703
www.principledtechnologies.com
When the assessment is complete, we know how to present the results to a
broad range of target audiences. We provide our clients with the materials they
need, from market-focused data to use in their own collateral to custom sales
aids, such as test reports, performance assessments, and white papers. Every
document reflects the results of our trusted independent analysis.
We provide customized services that focus on our clients’ individual
requirements. Whether the technology involves hardware, software, Web sites,
or services, we offer the experience, expertise, and tools to help our clients
assess how it will fare against its competition, its performance, its market
readiness, and its quality and reliability.
Our founders, Mark L. Van Name and Bill Catchings, have worked together in
technology assessment for over 20 years. As journalists, they published over a
thousand articles on a wide array of technology subjects. They created and led
the Ziff-Davis Benchmark Operation, which developed such industry-standard
benchmarks as Ziff Davis Media’s Winstone and WebBench. They founded and
led eTesting Labs, and after the acquisition of that company by Lionbridge
Technologies were the head and CTO of VeriTest.
Principled Technologies is a registered trademark of Principled Technologies, Inc.
All other product names are the trademarks of their respective owners.
Disclaimer of Warranties; Limitation of Liability:
PRINCIPLED TECHNOLOGIES, INC. HAS MADE REASONABLE EFFORTS TO ENSURE THE ACCURACY AND VALIDITY OF ITS TESTING, HOWEVER,
PRINCIPLED TECHNOLOGIES, INC. SPECIFICALLY DISCLAIMS ANY WARRANTY, EXPRESSED OR IMPLIED, RELATING TO THE TEST RESULTS AND
ANALYSIS, THEIR ACCURACY, COMPLETENESS OR QUALITY, INCLUDING ANY IMPLIED WARRANTY OF FITNESS FOR ANY PARTICULAR PURPOSE.
ALL PERSONS OR ENTITIES RELYING ON THE RESULTS OF ANY TESTING DO SO AT THEIR OWN RISK, AND AGREE THAT PRINCIPLED
TECHNOLOGIES, INC., ITS EMPLOYEES AND ITS SUBCONTRACTORS SHALL HAVE NO LIABILITY WHATSOEVER FROM ANY CLAIM OF LOSS OR
DAMAGE ON ACCOUNT OF ANY ALLEGED ERROR OR DEFECT IN ANY TESTING PROCEDURE OR RESULT.
IN NO EVENT SHALL PRINCIPLED TECHNOLOGIES, INC. BE LIABLE FOR INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES IN
CONNECTION WITH ITS TESTING, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. IN NO EVENT SHALL PRINCIPLED TECHNOLOGIES,
INC.’S LIABILITY, INCLUDING FOR DIRECT DAMAGES, EXCEED THE AMOUNTS PAID IN CONNECTION WITH PRINCIPLED TECHNOLOGIES, INC.’S
TESTING. CUSTOMER’S SOLE AND EXCLUSIVE REMEDIES ARE AS SET FORTH HEREIN.
Dell PowerEdge C5220: Hadoop performance
A Principled Technologies test report 13