Page 1
AAAM (Automated Availability Manager) agents,
334abstraction in vCenter design, 439AC power supply, 101–102acceptance levels in VIBs, 25access control
LUN masking, 216network, 368–371VMs, 93
active/active SAN arrays, 225Active Directory authentication, 47–48Active Directory Lightweight Directory Services
(AD LDS), 84, 84Active Directory Services Interfaces (ADSI) Edit
tool, 84, 85active/passive SAN arrays, 225AD Authentication Proxy tool, 48AD LDS (Active Directory Lightweight Directory
Services), 84, 84Add-DeployRule command, 37Add-EsxSoftwareDepot command, 27, 37Add-EsxSoftwarePackage command, 27addresses
IP. See IP addressesMAC, 284
admission controlHA, 338–341, 339–340resource pools, 319
ADSI (Active Directory Services Interfaces) Edit tool, 84, 85
advanced VM options, 262–263, 262affi nity rules
DRS, 290, 296–297, 324–327, 324–325FT, 352HA, 344VMFS-5 volumes, 240–241
agents in ESXi, 23–24alarms
HA, 344vCenter Server, 77, 391, 391
alerting operators, 400alignment of disk partitions, 202, 287–288
allocation models for vDCs, 471allocation pools for vDCs, 471–473, 471Allow Overlapping External Networks option,
458, 460Allow Publishing Catalogs to All Organizations
option, 461Allow the User to Specify option, 56altbootbank partition, 28ALUA (asymmetric logical unit access), 226–228,
228AlwaysOn Availability Groups, 301–302anti-affi nity rules, 296, 324, 352antivirus software optimization, 290antivirus storms, 381appliances
virtual, 73, 294–295VSA, 209–211
applicationsavailability monitoring, 397interoperating, 128management layer, 64–69monitoring, 297, 342–345, 342–343rollout benefi ts, 3
archives in ESXi design, 24arrays
compression, 197SAN, 225SATP, 225–226thin-provisioning, 195
assembling design, 15–16, 15assessing environment, 13–14assumptions, 5, 10asymmetric logical unit access (ALUA), 226–228,
228Atomic Test & Set locking, 231attributes for VMs, 264auditing, 385–386authentication
CHAP, 176–177, 219ESXi deployment, 47–48vCLI, 66, 67
Auto Deploy featurecomponents and process, 36–37deployment modes, 37–38
IndexNote to the Reader: Throughout this index boldfaced page numbers indicate primary discussions of a topic. Italicized page numbers indicate illustrations.
bindex.indd 479bindex.indd 479 2/20/2013 4:13:22 PM2/20/2013 4:13:22 PM
COPYRIG
HTED M
ATERIAL
Page 2
480 | AUTOMATED AVAILABILITY MANAGER (AAM) AGENTS • CAMPUS CLUSTERS
deployment scaling, 40description, 32ESXi, 34infrastructure, 35–36recommendations, 38stateful installs, 30
Automated Availability Manager (AAM) agents, 334
Automatic DPM mode, 328automation
cloud, 428DPM, 328, 328DRS, 321, 321vCloud Director, 429, 431VMFS-5 volumes, 239–240
availability, 161–162, 162HA. See High Availability (HA)I/O virtualization, 158management layer design, 76–82, 78management traffi c, 162–164, 163storage
effi ciency, 183–185IP, 165–168, 166, 168shared, 212
VM, 295–296monitoring, 397third-party clustering, 298–301, 301traffi c, 164–165, 165vSphere, 296–298, 297
availability design principle, 9–10average utilization data, 402
Bbackups
local databases, 74point-in-time copies as, 291in security, 383vCloud, 477–478
balancingDRS loads. 319–324, 321–323NLB, 300–301VMFS-5 volumes, 238–239
ballooning, 111–112bandwidth
measuring, 197NFS, 221–222
bare-metal hypervisors, 19
bedding-in, 98best practices in design, 16binary translation (BT) virtualization, 108binding, port, 228BIOS confi guration, 122blade servers, 131–132
cons, 133–135pros, 132–133vs. rack servers, 136
block-level deduplication, 196block sizes in VMFS, 193Block Zeroing primitive, 231blocked-based databases, 231–232Boot Options settings, 261, 261boot storms, 290bootbank partition, 28booting ESXi installer, 24British Thermal Units (BTUs), 103browser-based tools, 54, 55BT (binary translation) virtualization, 108BTUs (British Thermal Units), 103BU (business unit) networks, 460bundled databases, 73burn-in of server hardware, 123bus I/O in scale-out, 126–127bus sharing in SCSI, 277business continuity planning, 383business costing inventory structure, 310business function inventory structure, 310business unit (BU) networks, 460business unit ownership, 401–402BusLogic parallel controllers, 276BusyBox environment, 65
CCAB (Cluster Across Boxes), 299–300, 301cabling
10GbE, 157blade servers, 132network, 142in scale-up, 126
cachescontroller, 203–206deduplication, 204pre-fetch, 204stateless, 34–35, 37
campus clusters, 334
bindex.indd 480bindex.indd 480 2/20/2013 4:13:22 PM2/20/2013 4:13:22 PM
Page 3
CANNOT PUBLISH CATALOGS OPTION • CONFIGURATION PARAMETERS OPTION | 481
Cannot Publish Catalogs option, 462capacity
cloud, 429CPU, 109effi ciency, 185–186memory, 116–117monitoring, 397overview, 183planning, 389
change in, 389–390in design, 400–408sample design, 416, 424summary, 408–409
storage. See storage capacityvCloud Director, 429
CAPEX (capital expenditure) costs, 186catalogs
cloud, 428vCloud, 461–464, 461–462vCloud Director, 429, 431
Category 6A cabling, 142CBRC (content based read cache), 206CD/DVD drives, 255, 256CDP (Cisco Discovery Protocol), 170–171cell design for vCloud, 435–437, 436central database servers, 74central management
hypervisors, 54–56, 55Switches and distributed vSwitches, 153
centralized log collection, 386certifi cates
ESXi deployment, 45SSL, 94
chain lengths for VMs, 474change
in capacity planning, 389–390managing, 378–379
CHAP (Challenge-Handshake Authentication Protocol), 176–177, 219
chargebackcloud, 429vCloud Director, 429
child resource pools, 316chipsets
motherboards, 118servers, 160
CIB (Cluster in a Box), 299–300, 301CIM (Common Information Model)
brokers, 23, 99–100hardware monitoring, 56–57
CiRBA tool, 401Cisco Discovery Protocol (CDP), 170–171Cisco UCS Servers, 138client-connected USB devices, 258clones of VMs, 290–291cloud computing
models, 136–137risk, 383–385vs. server virtualization, 428–429vCloud. See vCloud design
Cluster Across Boxes (CAB), 299–300, 301Cluster in a Box (CIB), 299–300, 301clusters, 221
FT, 348, 353in HA. See High Availability (HA)inventory, 309Microsoft application, 301–302overallocating, 316remote databases, 74size, 288, 314–315stretched, 334, 346–347third-party, 298–301, 301vCenter, 311–315VMFS-5 volumes, 235–236VSA, 209vSphere storage, 243, 245–246
CNAs (converged network adapters), 217co-scheduling of CPU, 107co-stop metric, 266COM (serial ports), 257, 257command-line access to hosts, 365–368, 366, 368Common Information Model (CIM)
brokers, 23, 99–100hardware monitoring, 56–57
community PVLANs, 150Compact mode for Edge gateways, 475compatibility matrix, 59compatibility of vCenter Server, 82compliance
Host Profi les, 56, 385–386Storage Profi les, 244
compressionarray, 197memory, 112
computing needs for server hardware, 99–100Confi guration Parameters option, 263
bindex.indd 481bindex.indd 481 2/20/2013 4:13:22 PM2/20/2013 4:13:22 PM
Page 4
482 | CONNECTIVITY • DATACENTER LICENSE
connectivityin design, 311physical, 142
consistencyclusters for, 312PCI slots, 120–121server hardware, 100–101
Console Operating System (COS), 21consolidation
benefi ts, 3vCenter Server, 71
consolidation ratios, 405constraints in design, 3, 5, 413consumable resources in vCloud, 437–438, 438content based read cache (CBRC), 206contingency plans for migration, 42controllersIOPS effects, 200–201, 203–206
SCSI, 276–277cluster settings, 300VMs, 255
converged hardware, 138–139converged network adapters (CNAs), 217cooling
blade servers, 134requirements, 101–103, 102in scale-up, 126
cores, enabling, 122Cores per Socket setting, 267COS (Console Operating System), 21costs
blade servers, 133CAPEX, 186as network protocol selection factor, 224in scale-up, 125storage, 183
counters, performance, 269, 392–393, 393CPU to memory design ratio, 129–130CPUID mask options, 269–270CPUs
capacity, 109Cores per Socket setting, 267CPUID mask options, 269–270design overview, 265–267, 267EVC feature, 313hot-plugging, 267–268HT Sharing and scheduling affi nity, 270limits, 269
multicore and scheduling, 107optimizing, 289–290performance counters, 269reservations, 268–269resources, 268in scale-up, 124servers, 96–97, 107shares, 268utilization monitoring, 397vCenter Server, 91vCPUs, 107–109, 253virtualization, 269
credentialsPowerCLI, 67–69vCLI, 66, 67
cross-host vMotion, 279–280, 320Current Host Load Standard Deviation setting,
323custom attributes for VMs, 264customized images in ESXi, 25–27
DDAGs (database availability groups), 301–302DAS (direct attached storage), 181das.confi g.fdm.isolationpolicydelaysec setting,
344das.failuredetectiontime setting, 163–164, 344das.iostatsinterval setting, 342das.isolationaddress setting, 163, 346das.isolationaddress0 setting, 345das.isolationshutdowntimeout setting, 337das.maxftvmsperhost setting, 352das.SlotCpuInMHz setting, 339das.SlotMemInMB setting, 339das.usedefaultisolationaddress setting, 345data deduplication, 195–196data protection, 381–383data source name (DSN) entries, 64, 64data transfer in cloud computing, 384–385database availability groups (DAGs), 301–302databases
local vs. remote, 73–75protecting, 80–81Update manager, 64, 64vCenter Server, 61, 87, 89–90, 90vCloud, 438–439
Datacenter license, 286
bindex.indd 482bindex.indd 482 2/20/2013 4:13:22 PM2/20/2013 4:13:22 PM
Page 5
DATACENTERS • DISTRIBUTED RESOURCE SCHEDULING | 483
datacenters. See vCenter ServerDatastore Disk Overallocation % trigger, 194Datastore Disk Usage % trigger, 194datastores
heartbeats, 343, 343size, 192–193VMFS-5 volumes, 235–236vSphere storage, 243–246
DAVG tool, 206DCUI (Direct Console User Interface), 22
description, 23hypervisors, 52, 52shell access, 65, 366
Debugging and Statistics option, 262Dedicated Failover Hosts admission control
policy, 341dedicated storage switches, 219deduplication
caches, 204data, 195–196
defragmentation of fi les, 288–289Dell servers, 139departmental inventory structure, 310dependent mode disks, 275–276deployment
ESXi. See ESXi hypervisorsvCenter Server, 73
depth, queue, 201–202design, 1
assembling, 15–16, 15best practices, 16capacity. See storage capacitydocumenting, 16–17environment assessment, 13–14facets, 5–9, 6–9factors overview, 144functional requirements, 2–4, 2, 4, 11–13implementing, 17network. See networksoverview, 1–5, 2, 4principles, 9–11sample. See sample designstorage, 182–183summary, 17vCloud. See vCloud design
destinations for ESXi Installable, 32direct attached storage (DAS), 181direct-connected networks, 464
Direct Console User Interface (DCUI), 22description, 23hypervisors, 52, 52shell access, 65, 366
direct vApp networks, 467, 468DirectPath I/O technique
description, 97, 120vNICs, 159–161, 282
disablinghardware, 259interleaving. 122shell, 366–367, 366
disaster recovery (DR)DPM, 330vCloud, 438, 477–478
Disaster Recovery/Business Continuity (DR/BC) benefi ts, 3
discovery of iSCSI targets, 219discovery protocols, 170–171diskpart.exe tool, 288disks
alignment, 202, 287–288clustering, 298–301, 301confi guration settings, 122–123ESXi, 27–29, 28IOPS, 197–199latency measurements, 197modes, 275–276optimizing, 289–291RAID. See Redundant Array of Independent/
Inexpensive Disks (RAID) technologiestiering, 204–205types, 274–275vCenter Server, 91VMs, 193–194, 255, 273–276, 273
Disregard Setting option, 56Distributed Management Task Force (DMTF)
standards group, 295distributed power management (DPM), 319, 327
automation levels, 328, 328host options, 329impacts, 329–330requirements, 327–328as server hardware selection factor, 96uses, 330–331
distributed resource scheduling (DRS), 319affi nity rules, 290, 296–297, 324–327, 324–325automation levels, 321, 321
bindex.indd 483bindex.indd 483 2/20/2013 4:13:22 PM2/20/2013 4:13:22 PM
Page 6
484 | DISTRIBUTED VSWITCHES • ESXI HYPERVISORS
DPM. See distributed power management (DPM)
effi ciency, 323–324and FT, 352load balancing, 319–324, 321–323load requirements, 320vCenter Server failure effect on, 76VM options, 321–322, 322VMFS-5 volumes, 236–242
distributed vSwitches, 152–154DMTF (Distributed Management Task Force)
standards group, 295DMZ, 371–373, 372
fully collapsed, 374, 374partially collapsed, 373–374, 373separation of storage, 374–375
Do Not Reserve Failover Capacity admission control policy, 341
documentationdesign, 16–17reviewing, 12
downstream decisions, 15, 15downtime, 44, 183–185DPM. See distributed power management (DPM)DR (disaster recovery)
DPM, 330vCloud, 438, 477–478
DR/BC (Disaster Recovery/Business Continuity) benefi ts, 3
drivers, vNICs, 281–284DRS. See distributed resource scheduling (DRS)DRS-only clusters, 300DSN (data source name) entries, 64, 64Dump Collector service, 46dvSwitches (vSphere Distributed Switches), 147,
152, 414, 454dynamic discovery of iSCSI targets, 219
EE1000 vNICs, 281EC2 (Elastic Compute Cloud) model, 137EDA (ESX Deployment Appliance), 32Edge gateways, 465–467, 466–467, 475, 476Edge Virtual Bridging/Virtual Ethernet Port
Aggregator (EVB/VEPA), 145effi ciency
DRS, 323–324storage, 183–186
eG Innovations tools, 395eight NICs, design scenario for, 179, 179Elastic Compute Cloud (EC2) model, 137elasticity in planning, 389elections for HA hosts, 333–334Embedded version of ESXi deployment, 33–34enabling
FT, 351lockdown mode, 367–368, 368sockets, 122
end users in vCloud, 466–470, 467–470Enhanced vMotion Compatibility (EVC), 108, 269,
313, 348environment assessment, 13–14EPT (Extended Page Tables), 109equipment in inventory structure, 311EST (external switch tagging), 284ESX, 19–20ESX Deployment Appliance (EDA), 32ESX System Analyzer tool, 43esxcfg-nas command, 67esxcfg-nics command, 66esxcfg-vmknic command, 66esxcli command, 227esxcli-info command, 39ESXCLI toolkit, 24ESXi hypervisors, 19
command-line access to hosts, 365–368, 366, 368
concept, 21–22deployments
Auto Deploy infrastructure, 36–38comparing, 38–41, 39Embedded, 33–34hardware requirements, 27image location, 40–41Installable, 30–33, 31scaling, 39–40Stateless, 34–36types, 27–28
designagents, 23–24components, 22–23customized images, 25–27disk layout, 27–29, 28overview, 22system images, 24–25
evolution, 19–22
bindex.indd 484bindex.indd 484 2/20/2013 4:13:22 PM2/20/2013 4:13:22 PM
Page 7
ESXI SHELL • FIREWALLS | 485
guest optimization, 289–291management tools
centralized management, 54–56, 55hardware monitoring, 56–57host-management, 51–54, 52–53logging, 56–57
migrating to, 42–45postinstallation design options, 45–51, 49in sample design, 413, 416–417in scale-up, 125selecting, 99–100upgrading, 41–42vCenter Server failure effect on, 77
ESXi Shell, 24, 44, 52–53, 53esxtop tool, 206, 392/etc/exports fi le, 176/etc/hosts fi le, 222/etc/vmware/esx.conf fi le, 226Ethernet
10GbE considerations, 156–158network cards, 119switch ports, 219
EVB/VEPA (Edge Virtual Bridging/Virtual Ethernet Port Aggregator), 145
EVC (Enhanced vMotion Compatibility), 108, 269, 313, 348
Execute Protection feature, 122existing business unit network use case, 460expandability as server hardware selection
factor, 98–99Expandable Reservation option, 318–319exports, NFS, 176, 221–222Extended Page Tables (EPT), 109Extended Statistics primitive, 232extents, VMFS, 190external networks in vCloud design, 456–461,
459–460external switch tagging (EST), 284extraneous hardware, 122
Ffacets, 1–2
operational, 8–9, 8organizational, 7–8, 7overview, 5, 6technical, 6–7, 6
failed HA hosts, 333–334
failover
for availability, 184–185
hosts for, 341path, 225
Failover Clustering, 298
failures
blade servers, 134
and scaling, 127–128
false positives in vCloud, 437FAST (fully automated storage tiering), 446
fast provisioning VMs, 473–474Fault Domain Manager (FDM), 26, 332, 454
fault tolerance (FT), 72, 347–348in availability, 298enabling, 351HA host monitoring, 336
hosts, 168impacts, 352–353recommendations, 352–354requirements and restrictions, 349–350as server hardware selection factor, 96
uses, 351–352versions, 348, 349vLockstep interval, 347–348, 349and vMotion, 175–176
fdisk tool, 288
FDM (Fault Domain Manager), 26, 332, 454
features as server selection factor, 96–97fenced vApp networks, 467, 468Fibre Channel (FC)
characteristics, 212–215host bus adapters, 119, 375
overview, 215–217SAN devices, 181–182
vCloud, 444–449, 445–446, 450Fiber Channel NPIV setting, 263
Fibre Channel over Ethernet (FCoE), 145
characteristics, 212–215CNAs, 119
HBAs, 375
overview, 217–218fi le-based databases, 232fi le defragmentation, 288–289fi le-level storage, 195–196fi rewalls, 375–377
physical, 376ports, 50–51
bindex.indd 485bindex.indd 485 2/20/2013 4:13:22 PM2/20/2013 4:13:22 PM
Page 8
486 | FIXED CONFIGURATION OPTION FOR HOST PROFILES • HARDWARE ASSIST FEATURES
virtual, 376VMware vShield, 377
Fixed Confi guration option for host profi les, 56Fixed policy for PSP, 226fl ash drives
effi ciency, 199host-based caches, 205in tiering, 204–205
fl exibilityin design, 389effi ciency, 186RDMs, 278in scale-up, 125
fl exible vNIC drivers, 281FlexPod architecture, 138FlexSE disks, 275fl ing tool, 37fl oppy drives, 256folders for inventory, 307four NICs, design scenarios for, 178, 178frames, jumbo, 282
with iSCSI, 219working with, 150–152
FreeNAS project, 212FT. See fault tolerance (FT)Full Copy for blocked-based databases, 231Full File Clone primitive, 232Full mode for Edge gateways, 475Fully automated DRS level, 321–322Fully Automated mode for VMFS-5 volumes,
239–240fully automated storage tiering (FAST), 446fully collapsed DMZ, 374, 374functional requirements
in design, 2–4, 2, 4gathering and defi ning, 11–13tools for, 406violating, 15
Ggeographical inventory structure, 310Get-Credential command, 68global permissions, 364–365, 365goals in sample design, 412GPT (GUID Partition Table), 28, 287groups, security, 93growth planning, 408GSX product, 20
guest software, 285defragmentation, 288–289disk alignment, 287–288licensing, 286–287optimizing, 289–291OS selection, 285–286time settings, 290–291
guestscustomization, 293vCenter Server failure effect on, 77
GUID Partition Table (GPT), 28, 287
HHA. See High Availability (HA)HA Advanced Runtime Info settings, 340, 340HA/DRS clusters, 300HA-enabled clusters, 300hard disks. See disksHard memory state, 114hard zoning, 216hardware
hypervisor requirements, 27, 100–101inventory, 401management tools, 106monitoring, 56–57, 401sample design, 413–414, 418–419server. See server hardwarein templates, 294vCenter Server, 91–92VMs, 250
basic, 251, 251CD/DVD drives, 255, 256CPUs, 253fl oppy drives, 256hard disks, 255maximums, 253–254memory, 255miscellaneous devices, 258–259network adapters, 255ports, 257–258, 257removing and disabling, 259SCSI controllers, 255versions, 251–252video cards, 256–257, 256VMCI devices, 257
hardware assist featuresCPU enhancements, 109CPU virtualization, 108
bindex.indd 486bindex.indd 486 2/20/2013 4:13:23 PM2/20/2013 4:13:23 PM
Page 9
HARDWARE-ASSISTED CPU VIRTUALIZATION • INACTIVE AND IDLE VM MONITORING | 487
enabling, 122memory mapping, 110–111MMU enhancements, 109
hardware-assisted CPU virtualization (HV), 108Hardware-Assisted Locking, 231
hardware-based licenses, 287
hardware compatibility list (HCL), 22, 99, 105–106
hardware iSCSI initiators, 218–219, 227HDS servers, 139
head LUNs, 191
heads, 200
heartbeats
datastores, 343, 343failed hosts, 333–334
vCenter Server, 78–79, 78High Availability (HA), 331
admission control, 338–341, 339–340cluster protection, 78, 80–81failover, 296
failure detection, 333–334
fault tolerance. See fault tolerance (FT)
host monitoring, 335–338, 335–336hosts, 333–334impacts, 344recommendations, 344–345requirements, 331–332stretched clusters, 346–347vCenter failure effects on, 78vCloud, 438
VM and application monitoring, 342–345, 342–343
VM options, 336–338, 336vSphere, 78, 332–334
High memory state, 114–115
Host-Affi nity rules, 287
host-based fl ash cache, 205host-connected USB devices, 258hostd daemon, 23
hosted hypervisors, 19
hostnames
ESXi deployment, 45NFS, 222
hosts
and availability, 296, 397
certifi cates, 45cluster designs, 245–246command-line access to, 365–368, 366, 368
DPM options, 329
fault tolerance, 350, 352–354
HA, 333–334, 337
inventory, 309
isolation, 337
management tools, 51–54, 52–53memory usage, 110
monitoring, 335–338, 335–336profi les, 54–56, 312, 385–386
redundancy, 161–163, 162–163in scale-out, 126–127
sizing, 130–131, 130swapping, 112–113
vCenter Server failure effect on, 77
virtual machine traffi c, 164–165, 165vMotion interface, 168
vSphere fault tolerance, 168
hot clones, 291
hot-plugging
CPUs, 267–268
disabling, 351
memory, 272
HP servers, 138
HV (hardware-assisted CPU virtualization), 108
HyperThreading (HT) feature
description, 107
enabling, 122
HT Sharing, 270
hypervisor bypass, 159
hypervisors. See ESXi hypervisors
IIaaS (Infrastructure as a Service), 136–137iBFT (iSCSI Boot Firmware Table) format, 32IBM servers, 139idle memory tax (IMT), 115, 271iGroups, 216iLO (Integrated Lights Out), 327Image Builder tool, 25–27, 36images
customized, 25–27location, 40–41profi les, 25–26system, 24–25
implementing design, 17IMT (idle memory tax), 115, 271inactive and idle VM monitoring, 405
bindex.indd 487bindex.indd 487 2/20/2013 4:13:23 PM2/20/2013 4:13:23 PM
Page 10
488 | INDEPENDENT NONPERSISTENT DISKS • ITIL
independent nonpersistent disks, 276
independent persistent disks, 276
Independent Software Vendors (ISVs), 326, 391,
395
infl ating balloon, 112
Information Technology Infrastructure Library
(ITIL), 378
Infrastructure as a Service (IaaS), 136–137
infrastructure management clusters, 437, 438Infrastructure Navigator, 302–303, 302initial implementation scenario, 460
initiators, iSCSI, 218–219, 226–227, 228inline deduplication, 196
Installable ESXi version, 30–33, 31instrumented design, 390
Integrated Lights Out (iLO), 327
Intelligent Platform Management Interface
(IPMI), 327
interactive installs, 30–31
interdependencies, 5
interface effects on IOPS, 200
interleaving, disabling, 122
internal networks for vCloud, 464, 464interoperability
applications, 128
in manageability, 169
vCenter Server, 82
interrupts
coalescing, 284
timing, 348
interviewing individuals, 12–13
inventory
monitoring, 401
structure, 305–311, 305, 307–308, 311vCenter Server, 61
I/O
blade servers, 134
card setup, 122–123
latency, 238
ports, 103–104
servers, 97, 119
virtualization, 158
VM levels, 192
I/O Imbalance Threshold setting, 238
I/O Load Balancing Invocation Interval setting,
238
Iometer tool, 206
IOPS factorscaches, 203–206calculating, 197–199controllers, 200–201disks, 197–199interface, 200measuring, 186, 197, 206–207partition alignment, 202queuing, 201–202RAID, 199–201, 199SIOC, 203tiering, 204–205transport, 201VMs, 203workload, 202write coalescing, 203
iostat tool, 207IP addresses
conventions, 169–170ESXi deployment, 45NFS, 222vCloud, 458
IP storage, 165–168, 166, 168network traffi c security, 176–177performance, 173teaming options, 154
IPMI (Intelligent Platform Management Interface), 327
iSCSI, 154–155characteristics, 212–215HBAs, 119initiators, 218–219, 226–227, 228IP storage, 165–168, 166, 168multipathing, 228network traffi c security, 176–177overview, 218–221
iSCSI Boot Firmware Table (iBFT) format, 32ISO storage requirements, 190isolated PVLANs, 149–150isolated vApp networks, 468, 469isolation
HA events, 334host, 337for security, 93vCloud Director, 431
ISVs (Independent Software Vendors), 326, 391, 395
ITIL (Information Technology Infrastructure Library), 378
bindex.indd 488bindex.indd 488 2/20/2013 4:13:23 PM2/20/2013 4:13:23 PM
Page 11
JEOS • LSI LOGIC SAS CONTROLLERS | 489
JJeOS (Just enough OS), 285–286jumbo frames
description, 282with iSCSI, 219working with, 150–152
Just enough OS (JeOS), 285–286
KKAVG tool, 206key network components, 141
physical connectivity, 142software, 144traffi c types, 142–143
kickstart scripts, 31–32, 44
LLACP (Link Aggregation Control Protocol),
147–148large pages, 111large receive offl oad (LRO) feature, 283Last Time Exited Standby fi eld, 329latency
DirectPath I/O, 120disk, 197host-based fl ash cache, 205measurement, 185NUMA, 117VMFS-5 volumes, 238VMs, 263
Latency Sensitivity setting, 263LBT (load-based teaming), 229leases for vApps, 463Leave Powered On setting, 337–338legal issues in cloud computing, 384Let vCenter Pick option, 56libraries, templates, 292licensing
ALUA, 227ESXi deployment, 46inventory structure, 311ISVs, 326monitoring tools, 396in scale-up, 125software, 286–287vSphere, 104
limited user roles, 362, 362limits
CPUs, 269
memory, 272
resource pool settings, 318–319
VMFS capacity, 190–191
link aggregation
NFS with, 166–168, 166, 168physical switches, 145–148, 146–147
Link Aggregation Control Protocol (LACP),
147–148
Link-Layer Discovery Protocol (LLDP), 170–171
Linked Mode
security, 363–365, 363–365vCenter Server, 73, 82–86, 84–85
links, logging, 347
LISP (Locator/ID Separation Protocol), 180
LLDP (Link-Layer Discovery Protocol), 170–171
load balancing
DRS, 319–324, 321–323link aggregation, 148
NLB, 300–301
load-based teaming (LBT), 229
local databases vs. remote, 73–75
local security groups, 93
local storage, 118–119, 209, 212
local user permissions, 47
Locator/ID Separation Protocol (LISP), 180
lockdown mode
enabling, 367–368, 368ESXi deployment, 48–49, 49logging links, 347
logical unit numbers (LUNs), 187
data protection, 381
masking, 216
VMFS, 190–191
logs
centralized collection, 386
ESXi deployment, 46
tools, 56–57
long-distance vMotion, 371
Low memory state, 115
LPT (parallel ports), 257
LRO (large receive offl oad) feature, 283
LSI Logic Parallel controllers, 277
LSI Logic SAS controllers, 277
bindex.indd 489bindex.indd 489 2/20/2013 4:13:23 PM2/20/2013 4:13:23 PM
Page 12
490 | LUNS • MONITORING
LUNs (logical unit numbers), 187data protection, 381masking, 216VMFS, 190–191
MMAC addresses, 284Maintenance Mode for VMFS-5 volumes, 240Manage tab for Web client, 250manageability, 168
design principle, 10interoperability in, 169I/O virtualization, 158naming and IP conventions, 169–170
managementoverhead, 186remote, 106, 370tools
centralized management, 54–56, 55hardware monitoring, 56–57host-management, 51–54, 52–53logging, 56–57
traffi c, 143, 162–164, 163vCenter design, 439vCloud, 433–435, 435, 437–438, 438
management layer, 59design, 76
availability, 76–82, 78key decisions, 69–76manageability, 82–86performance, 86–92, 90recoverability, 92security. 92–94
PowerCLI, 67–69summary, 94vCenter Server, 59–61vCLI, 65–66vMA, 69vSphere Client and vSphere Web Client,
62–63VUM, 63–64, 64
Management Network port group, 210management networks
performance, 171–172, 172security, 174–175
Manual DPM mode, 328Manual DRS level, 321Manual Mode for VMFS-5 volumes, 240
mappingmemory, 110–111RDMs, 190, 277–279
masking LUNs, 216master boot record (MBR), 28, 287master HA hosts, 333–334MBps measurement, 183, 185, 197mean time between failures (MTBF), 184–185mean time to recover (MTTR), 184measuring
IOPs, 186, 197, 206–207storage performance, 197
memory, 110capacity, 116–117CPU to memory design ratio, 129–130FT, 352hot-plugging, 272limits, 272mapping, 110–111NUMA, 117–118, 272optimizing, 289–290overcommitment, 111–116, 116, 397for performance, 96–97reservations, 114–115, 271resources, 271in scale-up, 124usage, 110utilization monitoring, 397VMs, 255, 270–272, 270VSA, 210
message signal interrupts (MSI), 283metro clusters, 334Microsoft application clustering, 301–302Microsoft Clustering Service (MSCS), 298–301,
301Microsoft Network Load Balancing, 300–301migrating to ESXi, 42–45Mirror Mode in vMotion, 236MLAG (multiswitch link aggregation), 146–147,
146–147MLC (multi level cell) technology, 199MMU enhancements, 109monitoring
alerting operators, 400applications, 297, 342–345, 342–343building into design, 390clusters for, 312hardware, 56–57, 401
bindex.indd 490bindex.indd 490 2/20/2013 4:13:23 PM2/20/2013 4:13:23 PM
Page 13
MOST RECENTLY USED (MRU) POLICY • NETWORKS | 491
hosts, 335–338, 335–336item selection, 396–398sample design, 416, 424summary, 408–409thresholds, 398–399tools, 391–396, 391–394VMs, 297, 297, 342–345, 342–343
Most Recently Used (MRU) policy, 226motherboards, 118MPPs (Multipathing Plugins), 225MSCS (Microsoft Clustering Service), 298–301,
301MSI (message signal interrupts), 283MTBF (mean time between failures), 184–185MTTR (mean time to recover), 184multi level cell (MLC) technology, 199multicast mode in NLB, 301multicore CPUs, 107multipathing, 154–155, 225
ALUA, 226–228, 228NAS, 229plugin, 226SAN, 225–226, 226
Multipathing Plugins (MPPs), 225multiple sites in vCloud, 476–477multiswitch link aggregation (MLAG), 146–147,
146–147multitenancy
cloud, 428vCloud Director, 429–430
“must” rules in VM-Host affi nity, 326–327, 344Must run on hosts in group rule, 300
Nnames
conventions, 169–170vCloud organizations, 461, 461VMs, 263–264
NAP (Network Access Protection), 381NAPI (New API) feature, 283NAS (network-attached storage), 187
multipathing, 229vs. SAN, 221virtual, 209
Native Multipathing Plugin (NMP), 225Native Snapshots primitive, 232native VLANs, 149
Navigator tool, 302–303, 302Nehalem chips, 107Nested Paging Tables (NPTs), 111, 351NetApp/Cisco Flexpod, 138NetIQPlateSpin Recon tool, 401NetQueue support, 119Network Access Protection (NAP), 381network adapters
VMs, 255VSA, 210network-attached storage (NAS), 187multipathing, 229vs. SAN, 221virtual, 209
Network File System (NFS), 155–156characteristics, 212–215data protection, 382exports, 176, 221–222IP storage, 165–168, 166, 168network traffi c security, 176–177overview, 221–223VAAI for, 232vCloud, 435–437, 436
Network I/O Control (NIOC), 156Network Load Balancing (NLB), 300–301Network Time Protocol (NTP), 45, 290Networking view for inventory, 310networks
access control, 368–371design, 141
10GbE considerations, 156–158availability, 161–168, 162–163, 165–166, 168future, 180I/O virtualization, 158IP storage, 154iSCSI, 154–155jumbo frames, 150–152key components, 141–144manageability, 168–171naming and IP conventions, 169–170NFS, 155–156performance, 171–173physical switch support, 145–148, 146–147recoverability, 173–174scenarios, 177–179, 177–179security, 174–177server architecture, 160–161SR-IOV and DirectPath I/O, 159–161
bindex.indd 491bindex.indd 491 2/20/2013 4:13:23 PM2/20/2013 4:13:23 PM
Page 14
492 | NEW API • OVERALLOCATION
summary, 180VLANs, 148–150, 150vMotion interface, 168vSphere FT, 168vSwitches and distributed vSwitches,
152–154ESXi deployment, 45future virtualization, 180I/O factors, 119optimizing, 290pool decisions, 455–456protocols, 170–171
fi ber channel, 215–217NFS, 221–223
sample design, 414, 419–420, 421utilization monitoring, 397vApps, 466–470, 467–470vCloud, 456–461, 459–460, 464–466, 464–466
New API (NAPI) feature, 283New-Datastore command, 68New-DeployRule command, 37New-EsxImageProfi le command, 27New Virtual Machine wizard, 251NFS. See Network File System (NFS)NICs
in design scenarios, 177–179, 177–179hosts, 161–163, 162–163performance, 171–172vNICs, 280
DirectPath I/O, 159–161drivers, 281–284vCloud, 436–437, 436
9 values for availability, 183–185NIOC (Network I/O Control), 156NLB (Network Load Balancing), 300–301NMP (Native Multipathing Plugin), 225node interleaving
disabling, 122NUMA, 117
non-uniform memory architecture (NUMA)memory affi nity, 272overview, 117–118vNUMA, 266–267
nonvolume license agreement contracts, 286normal mode disks, 275notes for virtual machines, 264NPIV, 278NPTs (Nested Paging Tables), 111, 351
NTP (Network Time Protocol), 45, 290ntpd daemon, 23NUMA (non-uniform memory architecture)
memory affi nity, 272overview, 117–118vNUMA, 266–267
OOff DPM option, 328offl ine software depots, 26online software depots, 26Open VM Format (OVF) standard, 73, 295Openfi ler project, 212operating expenses (OPEX)
costs, 186power, 101
operating systems (OS)selecting, 285–286vCenter Server, 60, 75–76, 87
operational facets, 1, 2, 8–9, 8operators, alerting, 400OPEX (operating expenses)
costs, 186power, 101
optimizationcapacity, 406guests, 289–291vSphere computing environment, 80–81
Oracle databases, protecting, 80–81Oracle products, 138orchestration
cloud, 428vCloud Director, 429, 431
organizational direct-connected external networks, 460, 464
organizational facets, 1, 2, 7–8, 7organizational internal networks, 464organizations, vCloud
designing, 461–464, 461–462networks, 464–466, 464–466
orphaned VMs and VM resources, 406OS (operating systems)
selecting, 285–286vCenter Server, 60, 75–76, 87
outages, scheduled, 185overallocation
clusters, 316monitoring, 396
bindex.indd 492bindex.indd 492 2/20/2013 4:13:23 PM2/20/2013 4:13:23 PM
Page 15
OVERCOMMITMENT • PHYSICAL VCENTER SERVER | 493
overcommitmentmemory, 111–116, 116monitoring, 397thin-provisioning, 193
overhead, 186OVF (Open VM Format) standard, 73, 295
PP2V (physical to virtual)
clusters, 299heartbeat installation, 79
P2Ving VMs, 253PaaS (Platform as a Service) model, 137parallel ports (LPT), 257paravirtualization, 108–109parity disks, 188Partially Automated DRS level, 321–322Partially Automatic cluster setting, 300partially collapsed DMZ, 373–374, 373partitions
10GbE, 157alignment, 202, 287–288FC, 216HA, 334splitting, 273–274
passwordsPowerCLI, 68vCLI, 68
patchesclusters, 315hosts, 46VMs, 381
path failover, 225Path Selection Plugin (PSP), 225–226pay-as-you-go pools, 472–473, 472PCI bus, 119–121PCI devices, 258–259PCIe connectors and slots, 160PDL (Permanent Device Loss) codesPDUs (power distribution units), 102peak utilization data, 402per ms latency measurement, 185per-site permissions, 363–364, 363–364Percentage of Cluster Resources Reserved
admission control policy, 340–341perfmon tool, 206performance, 171
array compression, 196defragmentation for, 288–289effi ciency, 185FT, 352I/O virtualization, 158management layer, 86–92, 90management networks, 171–172, 172monitoring, 397as network protocol selection factor, 224NFS, 222as server selection factor, 97–98storage, 183, 197
IOPS. See IOPS factorsIP, 173shared, 212
thin-provisioning, 195vCenter Server charts, 392, 392vMotion, 172–173vNIC drivers, 284VSA, 210–211vSphere storage, 233–242, 237
performance counters, 269, 392–393, 393performance design principle, 10Permanent Device Loss (PDL) codespermissions
ESXi deployment, 47global, 364–365, 365per-site, 363–364, 363–364vCenter, 93–94, 360–363, 361–362
perspective in scaling, 127PFs (physical functions), 121PHD Virtual tool, 395physical compatibility mode RDM, 278physical connectivity, 142physical design, 6physical fi rewalls, 376physical functions (PFs), 121physical hardware-based licenses, 287physical network cable, 157physical switch support
jumbo frames, 150–152link aggregation, 145–148, 146–147, 166–168,
166, 168VLANs, 148–150, 150
physical to virtual (P2V)clusters, 299heartbeat installation, 79
physical vCenter Server, 70
bindex.indd 493bindex.indd 493 2/20/2013 4:13:23 PM2/20/2013 4:13:23 PM
Page 16
494 | PLANNING • QUEUING EFFECTS ON IOPS
planningcapacity. See capacityclusters for, 312
PlateSpin Recon tool, 401
Platform as a Service (PaaS) model, 137plug-ins for vCenter Server, 91Pluggable Storage Architecture (PSA), 225
point-in-time copies as backups, 291
policies
swapfi les, 313–314vCloud, 461–464, 461–462
pools
network pool decisions, 455–456resource, 315–319, 317vDCs, 471–475, 471–472, 475
port-based security, 370–371port binding, 228
port groups
network pools, 455
VSA, 210
port zoning, 216
portability of vCenter Server, 71–72
PortFast setting, 345
ports
fi rewall, 50–51server hardware, 103–104VMKernel, 166
VMs, 257–258, 257postinstallation design options, 45–51, 49power
blade servers, 132–133
DPM. See distributed power management
(DPM)
management options, 261in scale-up, 126
server requirements, 101–103, 102settings, 122
power distribution units (PDUs), 102
Power Off setting, 337–338
power supply units (PSUs), 101–103, 102PowerCLI tool, 54, 67–69PowerShell, 67
pre-fetch caches, 204preproduction checks for server hardware, 123previrtualization capacity planning, 401–405primary HA hosts, 334–335primary PVLANs, 149
principle of least privilege, 362
principles, design, 9–11priorities
inventory structure, 311
restart, 336–337, 344
private VLANs (PVLANs), 148–150, 150privileges. See permissions
processors. See CPUs
products
cloud, 428
vCloud Director, 429
profi les
host, 312, 385–386hypervisors, 54–56images, 25–26VM storage, 280vSphere storage, 243–245
projects in inventory structure, 311
promiscuous PVLANs, 149protocols
characteristics, 212–215choosing, 224–225iSCSI, 218–221network discovery, 170–171
Provider vDCs in vCloud
in design, 454–455, 470–472, 471–472logical side, 449–455, 450–451, 453–454physical side, 444–448, 444–449
PSA (Pluggable Storage Architecture), 225
PSP (Path Selection Plugin), 225–226PSUs (power supply units), 101–103, 102public Internet for vCloud, 458purpose of hypervisors, 100PuTTY tool, 65
PVLANs (private VLANs), 148–150, 150PVSCSI controllers, 277PXE environment
Auto Deploy, 36
booting, 30ESXi deployments, 39–40
QQUED tool, 206Quest tools, 395queue depth, 201–202queuing effects on IOPS, 201–202
bindex.indd 494bindex.indd 494 2/20/2013 4:13:23 PM2/20/2013 4:13:23 PM
Page 17
RACK SERVERS • ROUND ROBIN (RR) PSP | 495
Rrack servers, 135–136rack space, 101, 186RAID. See Redundant Array of Independent/
Inexpensive Disks (RAID) technologiesRAID write penalty, 200RAM. See memoryRAM-based storage cache, 205–206ramdisks, 29Rapid Spanning Tree Protocol (RSTP), 219Rapid Virtualization Indexing (RVI), 109raw device mapping disks (RDMs), 190, 277–279read-cache devices, 205receive-side scaling (RSS) feature, 283reclaiming memory, 111–116, 116reclamation for blocked-based databases,
231–232recoverability
design, 10–11, 173–174I/O virtualization, 158vCenter Server, 92
redundancyfor availability, 184–185hosts, 161–163, 162–163remote databases, 74–75in scale-out, 126server hardware selection factor, 98vCenter Server, 72, 78–79
databases, 80–81HA clusters, 78heartbeat, 78–79, 78, 81–82
Redundant Array of Independent/Inexpensive Disks (RAID) technologies
controller settings, 122IOPS effects, 199–201options, 187, 187RAID 0, 187, 199RAID 5, 188, 200RAID 6, 188–189, 200RAID 10, 188, 199–200RAID-DP, 189RAID-Z, 189storage rules, 189vCloud, 444–445, 445vendor-specifi c, 189
regular memory reclamation cycle, 115reliability of server hardware, 98
remote access cards, 123
remote administration, 367
remote console settings, 260, 260remote databases, 73–75
remote logging, 46
remote management, 106, 370
removable media
ESXi deployments, 40
ESXi Embedded, 33
Remove-EsxSoftwarePackage command, 27
removing hardware, 259
replication, 246–247
reservations
CPUs, 268–269
memory, 114–115, 271
resource pools, 318–319
vDC pools, 471–473, 472Reserve Space primitive, 232
resource pools, 315–317
admission control, 319
inventory, 309
settings, 317–319, 317vCenter Server failure effect on, 76–77
vDC, 471–473, 472resources
vCenter Server, 70
VMs
CPUs, 268
memory, 271
remote databases, 74–75
usage pattern monitoring, 401
responsibilities for vCloud, 437
restart priority, 336–337, 344
restores for local databases, 74
resxtop tool, 392
Retain IP/MAC Resources option, 458, 460reviewing documentation, 12
ring size in vNICs, 283
risks, 5
cloud computing, 382–385
scaled-up architecture, 127–128
vCenter Server, 71
roles
vCenter Server, 85–86
vCloud Direct, 429–430
root object in inventory, 306–307
Round Robin (RR) PSP, 226
bindex.indd 495bindex.indd 495 2/20/2013 4:13:23 PM2/20/2013 4:13:23 PM
Page 18
496 | ROUTED EXTERNAL NETWORKS • SELF-SERVICE PROVISIONING
routed external networks, 465–466, 466routed vApp networks, 468, 468RSS (receive-side scaling) feature, 283RSTP (Rapid Spanning Tree Protocol), 219runtime information for admission control, 340,
340runtime leases, 463RVI (Rapid Virtualization Indexing), 109
SSaaS (Software as a Service), 137sample design, 411
hypervisor selection, 413, 416–417monitoring and capacity planning, 416, 424networking confi guration, 414, 419–420, 421overview, 411–412security architecture, 415–416, 424server hardware, 413–414, 418–419shared storage confi guration, 414–415,
421–422summary, 425VM design, 415, 423VMware Datacenter, 415, 423–424vSphere management layer, 413, 417–418, 418
SANs. See storage area network devices (SANs)SAS in vCloud, 444–445, 445–446, 449–451, 450SATA drives in vCloud, 444–445, 445–446, 449,
450SATP (Storage Array Type Plugin), 225–226scalability
server hardware performance, 97vCenter Server, 82vCloud, 438
scale-up vs. scale-out for server hardware, 123–125
advantages, 125–127CPU to memory design ratio, 129–130host sizing, 130–131, 130perspective, 127risk assessment, 127–128size selection, 128–129
scalingESXi deployments, 39–40hypervisor requirements, 100rack servers, 135
scheduled outages, 185
schedulingCPU, 107NUMA, 117
scheduling affi nity, 270scratch partitions, 28–29, 46
screensavers, 289
scripts
ESXi Installable, 31–32
ESXi migration, 44
SCSI controllers
cluster settings, 300types, 276–277VMs, 255, 259
SDRS Rules tab, 263SE sparce (Space Effi cient) disks, 275
secondary HA hosts, 334–335secondary PVLANs, 149–150security, 174
auditing and compliance, 385–386change management, 378–379cloud computing, 383–385command-line access to hosts, 365–368, 366,
368data protection, 381–383DMZ, 371–375, 372–374fi rewalls, 375–377I/O virtualization, 158
importance, 357, 358IP storage network traffi c, 176–177management network, 174–175network access, 368–371NFS, 222
profi les, 49–50sample design, 415–416, 424separation of duties, 358–360summary, 387vCenter Linked Mode, 363–365, 363–365vCenter Server, 92–94, 360–363, 361–362vCloud Director, 431
VM traffi c, 175vMotion and FT traffi c, 175–176VMs, 379–381
security design principle, 11self-service provisioning
for cloud, 428
for vCloud Director, 429
bindex.indd 496bindex.indd 496 2/20/2013 4:13:23 PM2/20/2013 4:13:23 PM
Page 19
SEPARATING • SNMP HARDWARE MONITORING | 497
separating
DMZ storage, 374–375IP storage, 176management networks, 369–370
separation of dutiesmanagement network, 175in security, 358–360vCenter Server, 70
serial ports (COM), 257, 257server hardware, 95
10GbE, 157architecture, 160–161BIOS confi guration, 122blade servers, 131–135burn-in, 123cloud computing, 136–137component overview, 106computing needs, 99–100considerations, 95–96converged, 138–139cooling, 103CPUs, 107–109I/O ports, 103–104memory, 110–118, 116motherboards, 118network I/O, 119PCI bus, 119–121power, 101–103, 102preproduction checks, 123rack servers, 135–136rack space, 101sample design, 413–414, 418–419scale-up vs. scale-out, 123–131, 130selection factors, 96–99settings, 122–123storage, 118–119summary, 139–140UPSs, 103vendor selection, 104–106vSphere licensing, 104
server virtualization vs. cloud, 428–429Service Console, 21, 44–45service level agreements (SLAs), 183sfcbd daemon, 23shadow VMs, 474shared storage, 212
DRS load balancing, 320
sample design, 414–415, 421–422shares
CPUs, 268resource pool settings, 317–318
“should” rules in VM-Host affi nity, 326–327, 344showback
cloud, 429vCloud Director, 429
Shut Down setting, 337–338shutdown of VMs, 297, 297sibling resource pools, 316simultaneous multithreading (SMT), 107single initiator zoning, 216single-instance storage, 195–196single level cell (SLC) technology, 199single point of failures, blade servers as, 134Single Root I/O Virtualization (SR-IOV), 97, 121,
159–161, 282–284single sign-on (SSO)
vCenter Server, 61vCloud, 434
SIOC (Storage I/O Control) featureVMFS-5 volumes, 234–235VMs 203
Site Recovery Manager (SRM), 246–247sites, vCloud, 476–477SiteSurvey tool, 351six NICs, design scenario for, 178, 17864-bit hardware, 285size
clusters, 288, 314–315datastores, 192–193hosts, 130–131, 130in scaling, 128–129vCenter Server, 86–89VMs, 265
SLAs (service level agreements), 183slave HA hosts, 333–334SLC (single level cell) technology, 199slot size in admission control, 339–340SMT (simultaneous multithreading), 107snapshots
disk modes, 275–276storage requirements, 190vCenter Server, 71VMDKs, 275–276
SNMP hardware monitoring, 47
bindex.indd 497bindex.indd 497 2/20/2013 4:13:23 PM2/20/2013 4:13:23 PM
Page 20
498 | SOCKETS • STORAGE CAPACITY
socketsenabling, 122in scale-up, 124
Soft memory state, 114–115soft zoning, 216softswitches, 144software
guest. See guest softwarenetwork components, 144
Software as a Service (SaaS), 137software-based virtualization, 108software bundles, 26software depots, 26software (SW) initiators, 218–219, 228solid-state drive (SSD) disks, 197–198southbridge, 160Space Effi cient (SE sparce) disks, 275Space Utilization Difference setting, 238Spanning Tree Protocol (STP), 146sparce disks, 275SplitRX feature, 283splitting VM partitions, 273–274SPs (storage processors), 200–201SQL database protection, 80–81SR-IOV (Single Root I/O Virtualization), 97, 121,
159–161, 282–284SRM (Site Recovery Manager), 246–247SSD (solid-state drive) disks, 197–198SSH access, disabling, 366–367, 366SSL Certifi cates, 94SSO (single sign-on)
vCenter Server, 61vCloud, 434
stacking workloads, 402, 405standard builds, 291standardized IP addresses, 170standardizing data, 402standards, design, 16Starting Offset setting, 287startup of VMs, 296, 297state archives, 24stateful Auto Deploy mode, 37stateful installs, 30stateless Auto Deploy mode, 37stateless caching, 34–35, 37stateless hosts, 36Stateless versions, 34–36static discovery of iSCSI targets, 219
statistics, vCenter Server failure effect on, 77storage, 118–119, 181
availability monitoring, 397capacity. See storage capacityin deployment, 46design factors, 182–183effi ciency, 183–186local, 209, 212multipathing, 225–228, 226–227network protocols. See networksoverview, 181–182performance, 197
IOPS. See IOPS factorsIP, 173measuring, 197monitoring, 397
sample design, 414–415, 421–422shared, 212summary, 247–248VMs, 272–273, 273
Cross-Host vMotion, 279–280disk modes, 275–276disks, 273–276, 273profi les, 280RDMs, 277–279SCSI controllers, 276–277Storage vMotion, 279
vSphere. See vSphere storagestorage area network devices (SANs), 182
booting from, 40–41LUNs, 32multipathing, 225–226, 226vs. NAS, 221with RDMs, 278tiering, 204–205virtual, 209
Storage Array Type Plugin (SATP), 225–226storage capacity, 187
array compression, 197data deduplication, 195–196datastore size, 192–193monitoring, 397RAID options, 187–189, 187requirements estimates, 189–190thin-provision VM disks, 193–195VMFS
block sizes, 193limits, 190–191
bindex.indd 498bindex.indd 498 2/20/2013 4:13:23 PM2/20/2013 4:13:23 PM
Page 21
STORAGE DRS • TUNNELED INTERNET | 499
vSphere, 233–242, 237storage DRS
vCloud, 452VMFS-5 volumes, 236–242
Storage I/O Control (SIOC) featureVMFS-5 volumes, 234–235VMs 203
storage leases, 463storage presentation, 216Storage Profi les, 243–245Storage view for inventory, 309Storage vMotion, 279, 452store partitions, 28StormTracker tool, 395STP (Spanning Tree Protocol), 146stretched clusters
description, 448, 449HA, 334, 346–347
structural facets, 1, 2STS (Atomic Test & Set) locking, 231Summary tab for Web client, 250, 250support as server vendor selection factor, 105Swap File Location setting, 262swapfi les, 112–113
policies, 313–314storage requirements, 190
switched FC (FC-SW), 215switches
10GbE, 157in design, 152–154jumbo frames, 150–152link aggregation, 145–148, 146–147, 166–168,
166, 168network, 142VLANs, 148–150, 150
synchronization, 290–291syslog daemon, 23Sysprep tool, 293system images in ESXi, 24–25system partition, 28system worlds in ESXi, 23
Ttagging
VLAN, 284–285VMs, 264
tardisks, 29
Target Host Load Standard Deviation setting, 322targets of iSCSI, 219TCP segmentation offl oad (TSO), 282technical facets, 1, 2, 6–7, 6technologies as server vendor selection factor, 106templates
inventory, 309storage requirements, 190vCenter Server failure effect on, 77VMs, 292–294
10GbE considerations, 156–158test environments for change, 378–379testing
ESXi migration, 42–43monitoring tools, 396
thick provision disks, 274thin provision disks, 193–195, 231–232, 274–275third-party tools
clustering, 298–301, 301monitoring, 395–396
three-node clusters, 210thresholds
DRS load balancing, 323monitoring, 398–399
tickless timers, 259tiering of disks, 204–205time settings, 290–291timers, tickless, 259timing interrupts, 348top tool, 207, 401topology, network, 142total lockdown mode, 49tower models, 131traffi c
management, 162–164, 163network, 142–143, 176–177VM, 164–165, 165
training, blade servers for, 134Transparent Interconnection of Lots of Links
(TRILL) protocol, 145, 180transparent page sharing (TPS), 111, 329transport effects in IOPS, 201trend monitoring, 407TRILL (Transparent Interconnection of Lots of
Links) protocol, 145, 180trust in security, 382TSO (TCP segmentation offl oad), 282tunneled Internet, 458
bindex.indd 499bindex.indd 499 2/20/2013 4:13:23 PM2/20/2013 4:13:23 PM
Page 22
500 | TURBO MODE SETTINGS • VCENTER SERVER
Turbo Mode settings, 122Twinax cabling, 142two NICs, design scenarios for, 177–178, 177two-node clusters, 209type 1 hypervisors, 19type 2 hypervisor, 19
UUCS (Unifi ed Computing Systems), 138Ultimate Deployment Appliance (UDA), 32unicast mode in NLB, 301Unifi ed Computing Systems (UCS), 138uninterruptible power supplies (UPSs), 102–103universally unique identifi ers (UUIDs), 478untagged VLANs, 149Update Manager (VUM), 54, 63–64, 64, 87–89updating
templates, 293VMs, 381
upgradability as server selection factor, 98–99upgrading
ESXi, 41–42VMs, 252
UPSs (uninterruptible power supplies), 102–103uptime, 185USB controllers, 258use cases for vCloud Director, 430–433user-defi ned datastores, 243user permissions, 47user worlds, 23utilization monitoring, 397, 407–408Utilized Space setting, 237–238UUIDs (universally unique identifi ers), 478
VV2V (virtual to virtual) installation, 79VA (volt amperes), 102VAAI (vSphere APIs for Array Integration), 194,
230–232, 279VADP (vSphere APIs for Data Protection), 230VAMP (vSphere APIs for Multipath), 230vApps
description, 295leases, 463options, 263vCloud, 454, 454, 466–470, 467–470
/var/log directory, 57
VASA (vSphere APIs for Storage Awareness),
230–233, 243, 451, 451Vblocks, 138VCD-NI (vCloud Director Network Isolation-
Backed) network pools, 455–456VCE (Virtual Computing Environment) coalition,
138
vCenter Infrastructure Navigator application,
302–303, 302, 395
vCenter Operations Manager (vCOPs), 394–395,
394, 443
vCenter Server, 59–60, 154, 305alarms, 391, 391applications overview, 64–65Auto Deploy, 37
availability, 78–79, 78clusters, 311–315components overview, 60databases, 61, 73–75, 87, 89–90, 90design, 439–441, 440–442DRS. See distributed resource scheduling
(DRS)
in ESXi deployment, 46
HA. See High Availability (HA)
hardware resources, 91–92Heartbeat product, 78–79, 78, 81–82inventory service, 61inventory structure, 305–311, 305, 307–308,
311linked mode, 82–86, 84–85, 363–365, 363–365manageability overview, 82operating systems, 60, 75–76, 87performance charts, 392, 392permissions, 360–363, 361–362physical, 70plug-ins, 91recoverability, 92redundancy. See redundancy
resource pools, 315–319, 317sample design, 413, 423–424security, 92–94single sign on, 61sizing, 86–89summary, 94, 355vApps, 73vCloud Director, 436, 436
bindex.indd 500bindex.indd 500 2/20/2013 4:13:23 PM2/20/2013 4:13:23 PM
Page 23
VCLI • VIRTUAL MACHINE FILE SYSTEM | 501
virtual, 70–72vSphere Web Client, 61Windows-based, 72–73
vCLI (vSphere command-line interface), 44, 53–54, 65–67, 67
vCloud design, 427backup and disaster recovery, 477–478cell and NFS design, 435–437, 436cloud vs. server virtualization, 428–429databases, 438–439end users, 466–470, 467–470management
clusters, 440vs. consumable resources, 437–438, 438physical design, 442–444
management stack, 433–435, 435multiple sites, 476–477networks
external, 456–461, 459–460organizational, 464–466, 464–466pool decisions, 455–456vApp, 466–470, 467–470
organizations, catalogs, and policies, 461–464, 461–462
Provider vDCslogical side, 449–455, 450–451, 453–454physical side, 444–448, 444–449
vCenter design, 439–441, 440–442vCloud Director, 65
in physical management design, 442–443role, 429–430use cases, 430–433
vDCs, 470–475, 471–472, 474–476vCloud Director Network Isolation-Backed
(VCD-NI) network pools, 455–456vCloud Infrastructure vCenters, 439, 441, 442vCloud Resource vCenters, 439–441, 442vCOPs (vCenter Operations Manager), 394–395,
394, 443vCPUs (virtual CPUs), 107–109, 253vDCs (virtual datacenters), 437
designing, 454–455, 470–475, 471–472, 474–476
logical side, 449–455, 450–451, 453–454physical side, 444–448, 444–449
VDI (virtual desktop infrastructure), 3, 330VDSs (vSphere Distributed Switches), 147, 152,
414, 454
Veeam Monitor tool, 395vendor-specifi c images, 25vendor-specifi c RAID options, 189vendors
ISVs, 326, 391, 395server hardware, 104–106
versionsFT, 348, 349VM hardware, 251–252
vFabric Hyperic application, 395vFoglight tool, 395VFs (virtual functions), 121VGT (virtual guest tagging), 285VHV (virtualized hardware virtualization), 269VIB Author tool, 25VIBs (VMware Installation Bundles), 25vicfg-nics command, 66–67vicfg-vmknic command, 66, 151vicfg-vswitch command, 367video cards, 256–257, 256View Storage Accelerator feature, 206views for inventory, 305–306, 305violating functional requirements, 15virtual appliances, 73, 294–295virtual compatibility mode RDM, 278Virtual Computing Environment (VCE) coalition,
138virtual CPUs (vCPUs), 107–109, 253virtual datacenters (vDCs), 437
designing, 454–455, 470–475, 471–472, 474–476
logical side, 449–455, 450–451, 453–454physical side, 444–448, 444–449
virtual desktop infrastructure (VDI), 3, 330Virtual Extensible LANs (VXLANs), 145, 456virtual fi rewalls, 376virtual functions (VFs), 121virtual guest tagging (VGT), 285virtual LANs (VLANs), 148–150, 150
tagging, 284–285traffi c separation, 176
Virtual Machine Disk Format (VMDK), 190, 192, 241
Virtual Machine File System (VMFS)block sizes, 193capacity limits, 190–191datastores, 29storage for, 118
bindex.indd 501bindex.indd 501 2/20/2013 4:13:23 PM2/20/2013 4:13:23 PM
Page 24
502 | VIRTUAL MACHINE INTERFACE • VMCI
VMFS-3 volumes, 233–234VMFS-5 volumes, 233–238, 454
Virtual Machine Interface (VMI), 249Virtual Machine Monitor (VMM), 21, 23virtual machines (VMs), 249
access, 93anti-affi nity, 241availability, 295–296
Microsoft application clustering, 301–302monitoring, 397third-party clustering, 298–301, 301vSphere, 296–298, 297
clones, 290–291components overview, 249–250, 250CPU design, 265–272, 267, 270data protection, 383DRS options, 321–322, 322fast provisioning, 473–474fi rewalls in, 375–377FT, 350–351, 354guest software, 285–291HA options, 336–338, 336hardware. See hardwarehost traffi c, 168inactive and idle, 405inventory, 309I/O levels, 192IOPS effects, 203memory design, 270–272, 270memory usage, 110monitoring, 297, 297, 342–345, 342–343naming, 263–264network design, 280–285, 280notes, custom attributes, and tagging, 264options
Advanced General, 262–263, 262Boot Options, 261, 261General Options, 260, 260power management, 261remote console, 260SDRS Rules, 263VMware tools, 261, 261, 264
overview, 221partition splitting, 273–274resources. See resourcessample design, 415, 423
security, 175, 379–381sizing, 265storage. See storagesummary, 303swap space requirements, 190templates, 292–294thin-provision disks, 193–195traffi c, 143, 164–165, 165, 175updating, 381vApps, 295vCenter Infrastructure Navigator, 302–303,
302vCenter Server failure effect on, 77virtual appliances, 294–295
virtual NUMA (vNUMA), 117, 266–267Virtual Storage Appliance (VSA), 209–211virtual storage devices, 209virtual switch tagging (VST), 285virtual to virtual (V2V) installation, 79virtualization
capacity planning during, 405–408CPU, 108–109, 269
virtualized hardware virtualization (VHV), 269VKernel company tools, 395VLAN-backed network pools, 455vlance adapters, 281VLANs (virtual LANs), 148–150, 150
tagging, 284–285traffi c separation, 176
vLockstep interval, 347–348, 349vLockstep process, 347VM Communication Interface (VMCI), 257VM CPUs (vCPUs), 107–109, 253VM-Host affi nity rules, 325–327, 325, 344, 352VM Memory object, 392VM Processor object, 392VM Restart Priority setting, 336VM sprawl, 292VM Storage Profi les, 244VM-VM affi nity rules, 324–325, 352vMA (vSphere Management Assistant) tool
description, 69hypervisors, 53–54migration, 44remote administration, 367
VMCI (VM Communication Interface), 257
bindex.indd 502bindex.indd 502 2/20/2013 4:13:23 PM2/20/2013 4:13:23 PM
Page 25
VMDK • VSPHERE METRO STORAGE CLUSTER | 503
VMDK (Virtual Machine Disk Format), 190, 192,
241
VMFS. See Virtual Machine File System (VMFS)
VMI (Virtual Machine Interface), 249
vmkDiagnostic partition, 28
VMkernel
ESXi design, 22executives, 24
iSCSI, 155load balancing, 320
ports, 166
traffi c, 143VMM (Virtual Machine Monitor), 21, 23
vmmemtctl tool, 111–112vmname.vswp directory, 113
VMnix, 21
vMotion
cross-host, 279–280, 320
for DRS load balancing, 320and FT traffi c, 175–176hosts, 168performance, 172–173security, 371as server selection factor, 96–97Storage vMotion, 279vCenter Server failure effect on, 77
vCloud, 452
in vSphere5, 236VMs. See virtual machines (VMs)
vMSC (vSphere Metro Storage Cluster), 346
vmstat tool, 401
VMware availability, 78VMware Capacity Planner, 401
VMware Convertor tool, 253
VMware Endpoint security (EPSEC), 380
VMware Go service, 32VMware Installation Bundles (VIBs), 25VMware Tools, 47, 261, 261, 264VMware vCenter Infrastructure Navigator
application, 395
VMware vCenter Server. See vCenter Server
VMware vCenter Update Manager Sizing
Estimator, 88VMware vFabric Hyperic tool, 395
VMware Virtual Storage Appliance, 209–211
VMware vShield, 377VMware vSphere hypervisor term, 20
VMXNET drivers, 281
VMXNET 2 (Enhanced) drivers, 281VMXNET 3 (Enhanced) drivers, 281vNetwork distributed switches (vDSs), 152, 414
vNICs, 280DirectPath I/O, 159–161drivers, 281–284vCloud, 436–437, 436
vNUMA (virtual NUMA), 117, 266–267volt amperes (VA), 102
volumes in VMFS, 190
VMFS-3, 233–234VMFS-5, 233–236, 454
VOPS Server tool, 395
vpxa agent, 23
vRAM licensing, 104VSA (Virtual Storage Appliance), 209–211VSA-Back End port group, 210
VSA-Front End port group, 210
vServices options, 263vShield Endpoint, 380
vSphere
fault tolerance, 168HA, 78, 332–334
vSphere APIs for Array Integration (VAAI), 194,
230–232, 279
vSphere APIs for Data Protection (VADP), 230
vSphere APIs for Multipath (VAMP), 230
vSphere APIs for Storage Awareness (VASA),
230–233, 243, 451, 451vSphere Client
hypervisors, 51
overview, 62–63vSphere command-line interface (vCLI), 44,
53–54, 65–67, 67vSphere Distributed Switches (VDSs), 147, 152,
414, 454
vSphere Management Assistant (vMA) tool
description, 69hypervisors, 53–54migration, 44
remote administration, 367vSphere Metro Storage Cluster (vMSC), 346
bindex.indd 503bindex.indd 503 2/20/2013 4:13:23 PM2/20/2013 4:13:23 PM
Page 26
504 | VSPHERE STORAGE • ZONES
vSphere storage, 186, 229–230
APIs, 230–232
management, 242–247
performance and capacity, 233–242, 237vSphere Update Manager (VUM), 54, 63–64, 64,
87–89
vSphere Web Client, 54, 61–63
VST (virtual switch tagging), 285
vSwitches in design, 152–154
vswp fi les, 112–113VUM (vSphere Update Manager), 54, 63–64, 64,
87–89VXLAN Tunnel End Points (VTEPs), 145, 456VXLANs (Virtual Extensible LANs), 145, 456
WW (watts), 103W32Time tool, 290Wake On LAN (WOL), 327warranties of server vendors, 105watts (W), 103watts/IOPS measurement, 186WDDM (Windows Display Driver Model), 256
Web client, 54, 61–63, 250, 250whitebox servers, 105who-focused questions, 7–8, 7wide NUMA, 117Windows-based vCenter Server, 72–73Windows Display Driver Model (WDDM), 256WOL (Wake On LAN), 327working directory setting for swapfi les, 313workload effects on IOPS, 202workload mobility, 180World Wide Names (WWNs), 215World Wide Node Names (WWNNs), 215World Wide Port Names (WWPNs), 215worlds in ESXi design, 22–23write coalescing for IOPS, 203write-through caching, 205
ZZenoss tool, 395ZFS fi le system, 189Zip fi les, 26zones, security, 216, 369–370, 373–374, 373
bindex.indd 504bindex.indd 504 2/20/2013 4:13:23 PM2/20/2013 4:13:23 PM