Exam Number SK0-002
Number of Questions :80
Minimum Passing Score :615/900
Time Alloted 90 Minutes
Exam Objectives: Available at the CompTia Site here (registration required)
A full rack is 42 U
One U is 1.75 inches
therefore a full rack has 73.5 inches of useable space.
Never move a full rack. Always remove all equipment first.
Install the heaviest parts towards the bottom. ( ex: UPS units)
Racks typically have wheels, and most have stabilizer feet.
Standard rack mounting area size is 19 inches including the mounting strips. Most modern racks use a square hole pattern, in which can be mounted cage nuts or leveraged by equipment designed to insert into the square hole pattern.
Most equiment is designed to mount to a set of rails, which then attach to the rack. This allows for easy access to the equipment.
Pedestal Case: Similar in appearance to a wide full tower desktop case
A blade server is a server mounted on a card that fits in a special chassis or shell approx. 3u in height
1-reduced space needed
2-improved server management
More information: 19 Inch Rack Wiki
JBOD- Just a bunch of disks
Zero Channel Raid – integrated into the motherboard or a standalone card.
Raid 0 -stripe set
stripes without parity
Read Write performance increased
Not fault tolerant
Faster reads, slower writes then single disk
Fully redundant at disk level
High overhead (100%)
Splitting/Duplexing each disk has its own controller (more redundancy)
Disk striping with parity
3 Drives minimum
Disk space is n-1
Disk striping with 2 parity stripes
4 drives minimum
Disk space is n-2
Can survive loss of two disks
more computing overhead to generate parity information
Striped array over two mirrored disk
possible to survive two disk failure (depends on which two disks)
More information: Raid University
SCSI And Disk Notes
SCSI-1 uses an 8-bit bus and supports data rates of 5 MBps.
SCSI-2 adds optional 16-bit support, data rates of up to 20MBps, and new connectors (including the widely known 50-pin D-sub connector).
Wide SCSI uses a wider cable (68 pins) to support 16-bit transfers.
Fast SCSI uses an 8-bit bus, but doubles the clock rate to support data rates of 10 MBps.
Fast Wide SCSI uses a 16-bit bus and supports data rates of 20 MBps.
Ultra SCSI uses an 8-bit bus and supports data rates of 20 MBps.
Ultra Wide SCSI uses a 16-bit bus and supports data rates of 40 MBps.
SCSI-3 is a collection of related standards, including the Ultra, Ultra2, and faster standards.
Ultra2 SCSI uses an 8-bit bus and supports data rates of 40 MBps.
Wide Ultra2 SCSI uses a 16-bit bus and supports data rates of 80 MBps.
Ultra3 SCSI uses a 16-bit bus and supports data rates of 160 MBps.
Ultra320 uses a 16-bit bus and supports data rates of 320 MBps. It requires a new data-transfer mode with an 80 MHz free-running clock to eliminate ISI problems with the clock signal.
Fiber Optic Technology
There are two types of fiber optic cable:
Singlemode -used for long distances. Uses a laser as a light source. Up to 100 km without a repeater.
Multimode – up to 2km . Uses LED as a light source. Lower cost.
POF Plastic Optic cable is an alternate type of cable, using a plastic core. Better ability for bends and getting beat on.
Cables are constructed in layers. The glass core is surrounded by a film (cladding) that protects the fiber and reflects light inward. A buffer coating -which may be gel filled- gives the cable some flexibility, and this is surrounded by a braided kevlar sleeve for strength. Finally, a PVC jacket surrounds the bundle for protection purposes.
Media converters- A fiber optic media converter changes electrical signals to optical signals.
lighter then copper. Info at the speed of light. impervious to eletricity and moisture.
FCIP Fiber Channel over IP. Tunnels fiber frames over TCP/IP. Sometimes used in point to point connections between SANs.
N_Port stands for Node Port and located at the termination (Node) end.
F_Port stands for Fabric Port and is located at the switch end.
Fiber Channel supports three (3) topologies:
Point-to-Point: two N_Ports (such as a Host and RAID system) connected. Can serve as a high-speed interface between two points over long distances. The communication is full duplex so a 1 gigabit/sec link delivers 200MB/s of dedicated bandwidth.
Arbitrated Loop: Fiber Channel Arbitrated Loop (FC-AL) allows multiple connections without the aid of a switch. Here, L_Port can be connected to any loop with each port sharing the bandwidth. The connection is half-duplex which means once a port arbitrates and gains access to the loop, transmission of information is in one direction only. This effectively opens a point-to-point connection utilizing 1/2 of the available loop bandwidth.
FC-AL supports full-duplex data-transfer rates of 100MBps and is compatible with SCSI.
Switched Fabric and Loop: Uses a Fiber Channel Switch . A switch will also automatically configure the port for point-to-point (F_Port) or loop operation (FL_Port).
Fiber is good for the network backbone because it is immune to EMI and also can carry higher bandwidth over long distances.
Cold Swap: It means that you have to power down a server before the part is replaced.
Warm Swap: It means that you may have to stop I/O functions corresponding to the part that need to be replaced by using appropriate commands. The server need not be powered off.
Hot Swap: Hot Swap allows the faulty part to be replaced without interrupting the Server or introducing any interruption in the I/O procedures.
Requires at least two servers.
Improves the availability of the network.
Clustering allows for better scalability.
If one of the servers in a cluster fails, other server(s) in the cluster take over the entire load of the failed server without interruption to the services.
Full: All files are backed up, and the archive bit is cleared.
Incremental: A full backup is done then the changed files are backed up, and the archive bit is cleared.
Differential: A full backup is done, then all changed files since the last full backups are backed up, and the archive bit is not reset. Restore is done with the full
Differential backups take more time than incremental backups on a daily basis because they will back up the data that was changed since the last full (or incremental) backup. However, if you’re using differential, it takes only the last full backup and the last differential backup to completely restore a system. Using incremental, the last full backup and every incremental backup must be used to restore a system.
Blind Patching refers to applying service packs without first testing the deployment
A group of patches is sometimes called a service pack.
Disaster Recovery notes
Hot Site — A fully equipped and operational data processing facility- ready to go. Very expensive.
Cold Site — Conditioned space, possibly with communications, environmental controls and power. No live data.
Warm Site — Conditioned space with communications, environmental controls and power, Equipment is in place. Data may be near line or brought in via removeable media such as tape.
Adapter load balancing vs adapter teaming – Teaming allows creating a high bandwidth channel out of the adapters while load balancing distributes the load across available adapters.
Bus Mastering allows devices to transfer information between themselves without involving CPU or system memory. Multiple devices can take over the bus without locking eachother out.
NAS Network Attached Storage
SAN Storage Area Network
Generally speaking, a SAN deals with block level access and NAS is file level access.
iSCSI Internet SCSI – encapsulates SCSI commands and data into TCP/IP packets. Provides a good protocol for SAN using standard IP networks on Ethernet hardware.
TOE TCP Offload Engine – One way of handling the overhead of iSCSI’s TCP encapsulation
AoE ATA over Ethernet- ATA protocol embedded in Ethernet frames
HBA Host Bus Adapter
ESCON Enterprise Systems Connection
Out of Band vs In Band– out of band keeps the control out of the data path – lots of configuration required (software on each server, etc) In band is easier to manage and deploy but has a small amount of latency (lower TCO) and can be a single point of failure.
Storage Virtualization -unification of storage – create a single pool managed across varied platforms and locations. Creates a logical view called a resource map.
Trunking Multiple physical paths between switches function as one logical path
Switch Fabric switches interconnected in a mesh topology
Common Diagnostic Tools and Commands
IPCONFIG /Release, /Renew /all /registerdns etc
ARP Show and manipulate Adress Resolution Protocol Tables
NETSTAT IP stats
NBTSTAT NetBIOS stats and info
DOWN -shut down server
exit -exit from dos mode
MONITOR – view stats
Linux/Unix based servers
shutdown (f fast -h halt -r restart now)
wall Broadcast message
WOL Wake on Lan
Allows a system to be powered on remotely
Connected via 3 pin cable
Uses the Magic Packet (wake up system)
Server Room Security
Secure access to the server room and backup tapes
All doors should lock either by key of card
Server rack doors should lock
Remove trashcans from the server room (no need for cleaning personnel in there)
Examples of biometrics include retina scanning, fingerprint reading and palm scanning.
Limit access to backup tapes.
Always have a disaster recovery plan- a how-to for recovering from a critical event
The DR plan should include a complete inventory of all devices.
The DR plan should be tested yearly
Class C fire extinguishers are carbon dioxide.
Server Room Environmental
Tempature Range approx 68-71 degrees farenheit
Relative Humidity approx 50%
ESD Electro Static Discharge is caused by hot and dry conditions.
Watts=Current x Volts
Line Interactive– power always goes through the UPS providing a smooth signal.
AVR Automatic Voltage Regulation – supply is withing tolerance window (Buck/Boost)
EMI ElectroMagnetic Interference
SNMP Simple Network Management Protocol
MIB Management Information Base
SMI Structure of Management Information
OID Object Identifier – uniquely IDs a path to a managed object.
The management station is the console with which the network is manipulated. Agents run on the devices being managed.
Versions: SNMPv1 ,v2 ,v3
v3 added security and remote configuration operations.
SNMP default ports are 160,161
ECC Error Checking and Correction
Servers use ECC Ram because it can detect and correct errors on the fly.
Memory Interlacing Technique to alternate between ram banks
Buffering temporary storage between system and ram
SIMM single inline memory module
DIMM dual inline memory module
EDO Extended Data Out 72 Pin 32 bit
SDRAM Syncronous Dyanmic RAM 144 pin 64 bit
DDR SDRAM Double Data Rate SDRAM 184 pin 64 bit
DDR2 240 pin 64 bit
DDR2 1.8 volts, 240 pins
RDRAM RAMBUS 800 Mhz 184 pin, must use a terminator in unused channels.
Backups should be performed regularly, in accordance with the corporate disaster recovery plan.
Popular Backup Strategies
Grandfather- monthly tape ,stored offsite
Father weekly tape
Son 4 daily tapes
Every file is backed up. archive bit reset- offers the fastest restore, at the expense of time required to back up.
Backs up only files that have changed since the most recent Full back up was done. Resets the archive bit.
Progressive Incremental Assumes all backups, including the fisrt full, are incremental.
Backs up any files that are determined to have changed since the performance of the most recent full back up. Does not clear the archive bit.
Does not clear archive bit
Tape Drives And Tape Types
Tapes are slow, but portable and rugged. Good for secure offsite of backup.
QIC Quarter Inch Cartridge
Travan -somewhat backward compatible with QIC.
DAT Digital Audio Tape – small cartidges, SCSI only. Uses DDS cartridges
8MM – Standard 8mm, Mammoth, and AIT
Mammoth -more reliable drives
AIT Advanced Intelligent Tape Technology- uses MIC Memory in Cassette to speed reads.
ADR Advanced Digital Recording
DLT Digital Linear Tape
Uses SPR Symmetric Phase Recording for greater track density.
SDLT Super Digital Linear Tape
LTO Linear Tape Open (Ultrium)
Robotic Libraries – Swaps tapes automatically for greater capacity.
Tape Arrays – Based on RAID- can stripe data across multiple tapes.
HSM Hierarchical Storage Management- uses online, near-line and offline storage to sort storage based on use frequency. The Policy Repository guides the decisions about how to store data. Windows uses RSS Remote Storage Service to present the impression that all files are in a central location even though HSM is being used in the enterprise.
Common Services and Default Ports
DNS Domain Name System 53
POP3 Post Office Protocol 110
SMTP Simple Mail Transfer Protocol 25
SNMP Simple Network Management Protocol 160,161
NNTP Network News Transfer Protocol 119
FTP File Transfer Protocol 21
SSL Secure Sockets Layer 443
Protocols and File Systems
PPP Point to Point Protocol
PPTP Point ot Point Tunneling Protocol
L2TP Layer 2 Tunneling Protocol
SNA Systems Network Architecture
DHCP Dynamic Host Configuration
ICMP Internet Control Message Protocol
FCIP Fiber Channel over IP
iFCP Internet Fibre Channel Protocol
SAS Serial Attached SCSI
SATA Serial ATA
SMP Symmetric MultiProcessing
STP Spanning Tree Protocol
SSP Switch to Switch Protocol
iSCSI Internet SCSI
HIPPI HIgh Performance Parallel Interface
SMB Server Message Block
NFS Network File System
CIFS Common Internet File System
NSS Novell Storage Systems
PKI Public Key Infrastructure
IPX Internetwork Packet Exchange
SPX Sequenced Packet Exchange
DMI Desktop Management Interface
NTLM NT LAN Manager
IPMI Intelligent Platform Management Interface
ICMP Internet Control Message Protocol
Server+ is © Comp-Tia.