1532xix.fm Page 344 Monday, October 18, 2004 4:38 PM

1532xix.fm Page 345 Monday, October 18, 2004 4:38 PM

I N D E X

RSVP, 99–101 Numerics TE, 89–90 traffic-handing mechanisms, 81–83 802.11 networks, QoS, 276 work-conserving queuing, 93–96 agents, 146–147, 192 capture, 234 A embedded, 72 enabling, 148 ABR (available bit rate), 294 profiling, 39 Abstract Syntax Notation One (ASN.1), 147 SAA, 72, 158, 199 access. See also security agreements (SLAs) storage-area networks, 269 Cisco NetFlow, 73–74 transactions, 232 defining, 65–66 access control lists (ACLs), 81, 95 delivery, 66–67 accountability, delivery of service, 133 management, 67–68 accounting, 143 monitoring, 76–77 accounts, transactions, 230 profiling, 74–76 acknowledgments (ACKs), 60, 254 requirements, 68–73 ACLs (access control lists), 81, 95 alarms, 142 active agents, 192 algorithms Address Resolution Protocol (ARP), 236 queuing, 82, 91. See also queuing administration. See also management queue-servicing, 81 APM, 5 alignment, business/technical methodologies, 30–32 benefits of, 16 allocation end-to-end systems, 7 ports, 243 isolation, 7–8 QoS, 297 lifecycles, 9 resources, 80–81, 108–109 monitoring SLAs, 9 amalgamation, 62 overview of, 6 analysis sharing information, 8 business-impact, 124 CRM systems, 74 capacity planning, 183–186 oversight, 26 failure, 181 QPM, 109 modeling, 51 SLAs, 67–68 predictive performance, 15 system, 13 queries, 257–258 TCO, 25 root-cause, 142, 150 tools, 79 SLA compliance, 180 configuration, 83–84 systems, 77 DiffServ, 101–102 TCO, 25–27 FIFO, 92–93 thresholds, 49 IntServ, 98–99 triggers, 14 MPLS, 85–88 virtual environments, 176 networks resource allocation, 80–81 analyzers non-work–conserving queuing, 96–98 data, 204 provisioning, 83–84 network, 234 queuing, 91–92 profiling, 38

1532xix.fm Page 346 Monday, October 18, 2004 4:38 PM

346 any-to-any communication

any-to-any communication, 270 assessment APIs (application programming interfaces), 190 applications, 14 APMs (application performance management), demarcation points, 128–130 5, 127 new application impact, 52 benefits of, 16 assignment end-to-end systems, 7 associated applications, 24 isolation, 7–8 demarcation points, 128–130 lifecycles, 9 QoS, 134–137 monitoring SLAs, 9 ATM (Asynchronous Transfer Mode), 86 overview of, 6 authentication, 110 services, 132–134 authorized transactions, 230 sharing information, 8 automation application programming interfaces (APIs), 190 capacity planning, 185 application-aware probes, 158 polling, 146 application-development teams, 13 availability, 157 applications, 13, 29 defining, 65–66 analyzers, 38 delivery, 66–67 architecture, 217–219, 256 available bit rate (ABR), 294 assessment, 14 average bytes per frame, viewing, 241 bandwidth-bounded, 42, 59–61 baselining, 48 business-critical, 29 B capture points, 55 characterizing, 45 background classification, 40–42, 289–290 data, 177 coexisting, 28 traffic, 118, 177. See also traffic data flow, 49 backplane utilization, 160 delivery, 11, 61–63 balancing metrics, 154 deployment, 50 bandwidth, 42 end-to-end models, 284–297 application consumption of, 7 interactive, 217 increasing, 61 latency-bounded, 40, 58–59 optimizing, 59–61 mission-critical, 291 priority, 29 monitoring, 198, 205 profiling, 36 probes, 72 windows, 254 status, 43 base networks, building, 118 storage solutions, 272 base transactions, defining, 43, 230. See also threads, 40 transactions turns, 37, 243 baselining, 47–50, 155 web-based, 291 businesses, 122 workloads, 10 collection, 185 architecture environments, 43 applications, 217–219, 256 establishing, 287 delivery requirements, 221–222 infrastructure, 107–108 hub-and-spoke, 135 performance, 173–174 policy management, 109–112 technical baselines, 121 ARP (Address Resolution Protocol), 236 BC (committed burst), 135 ASN.1 (Abstract Syntax Notation One), 147 Be (excess burst), 135

1532xix.fm Page 347 Monday, October 18, 2004 4:38 PM

CLEC (competitive local exchange carrier) 347

benchmarking, 121 TCO, 25 benefits of TE, 90 APM, 16 validating, 179 TCO, 27 capacity Berkeley Standard Distribution (BSD), 204 increasing, 61 BGP (Border Gateway Protocol), 83 planning, 66, 174, 183–186 big bang deployments, 119–120 capturing big-picture view, 216 capture points, 55 bits, DE, 135 data, 54 Border Gateway Protocol (BGP), 83 traces, 218–219, 234–235 bottlenecks, identifying, 289 traffic, 218 boundaries, defining transactions, 248 transactions, 236 BPDU (bridge protocol data unit), 204 CAR (committed access rate), 81 branches, transactions, 232 case studies bridge protocol data unit (BPDU), 204 QoS, 259–265 broadcast traffic (etherStatsBroadcastPkts), 161 SLAs, 246–253 BSD (Berkeley Standard Distribution), 204 transactions BSM (business service management), 133 optimizing, 253–259 building profiling, 229–246 baseline environments, 43 categorizing business transactions, 230 baselines, 48 CBR (constant bit rate), 294 business applications, 7 CBWFQ (class-based weighted fair queuing), 81, 95 business baselines, 122 CDP (Cisco Discovery Protocol), 149 business criteria category (SLAs), 68–69 Cell Loss Priority (CLP), 135 business management, 150 cell switching, ATM, 86 business objectives, 132–134 central site (hub), 135 business requirements, 127 channel service unit/data service unit business service management (BSM), 133 (CSU/DSU), 295 business transactions characterizing applications, 45 optimizing, 253–259 CIFS (Common Internet File System), 271 profiling, 229–246 CIR (committed information rate), 135 business-critical applications Cisco Discovery Protocol (CDP), 149 defining, 22–24 Cisco IOS priority, 29–30 show commands, 201 requirements, 27–29 traffic-conditioning mechanisms, 136 TCO, 25–27 Cisco NetFlow, 31, 73–74 technical alignment, 30–32 Cisco Networking Service (CNS), 148, 213 business-impact analysis, 124 Cisco Tag Switching implementations, 86 bytes Cisco Works, 31 transfers, 245 class-based weighted fair queuing (CBWFQ), viewing, 241 81, 95 classification, 81, 84, 135–136 applications, 40–42, 289–290 C devices, 112 management systems, 141–151 calculations clean transactions, 235 measurements, 72 cleaning traces, 236 SPF, 90 CLEC (competitive local exchange carrier), 291

1532xix.fm Page 348 Monday, October 18, 2004 4:38 PM

348 CLI (command-line interface)

CLI (command-line interface), 109, 147, metrics, 250 199–204 QoS, 109, 297 client, network, and server (CNS), 148, 213 RSPAN, 164 clients SPAN, 163 overloading, 220 storage-area networks, 269–276 processing time, 252 TCO, 25–27 CLP (Cell Loss Priority), 135 thresholds, 49 CNS (Cisco Networking Service), 148, 213 validation, 180 CNS (client, network, and server), 148, 213 conflict-prevention tools, SLAs, 67 coexisting applications, 28 congestion, troubleshooting, 176 collection congestion-avoidance (WRED) procedures, 135 baselining, 185 connections key date points, 70 any-to-any communication, 270 methods, 72 bandwidth-bounded applications, 42 metrics, 157–165 critical switched ports, 71 points, 246–253 latency, 35 collisions, Ethernet, 161 WLANs, 276–284 command-line interface (CLI), 109, 147 consistency commands APM, 6 debug, 202–203 transaction captures, 236 frame-relay ip rtp header-compression, 263 constant bit rate (CBR), 294 get, 245 constraint-based routing, 90 ip rtp compression-connections, 264 consumption, monitoring, 10 ip rtp header-compression, 263 contingency plans, deployment, 121–124 show, 201 control consoles, reporting to, 192 show processes cpu, 204 conversation, 38 snmpwalk, 49 capture points, 235 committed access rate (CAR), 81 paths, 157, 247 committed burst (Bc), 135 conversational maps, 240 committed information rate (CIR), 135 COPS (Common Open Policy Service), 110 Common Internet File System (CIFS), 271 core devices, QoS, 113 Common Open Policy Service (COPS), 110 core networks, CPE measurements, 72 common sense, 179 correlation of events, 142 communication CPE (customer premises equipment), 72, 295 deployment, 124 CQ (custom queuing), 81, 94 directory systems, 110 credit, SLAs, 66 SLAs, 67 critical switched port connections, 71 comparing trace files, 237–246 CRM (customer relationship management), 74 competitive local exchange carrier (CLEC), 291 cRTP (Real-Time Transport Protocol header complex transactions, 230 compression), 263 compliance, SLA analysis of, 180 CSU/DSU (channel service unit/data service computing resources, forecasting, 10 unit), 295 Compuware Application Vantage, 234 custom queuing (CQ), 81, 94 configuration customer premises equipment (CPE), baselining, 48 72, 295 CLI, 109 customer relationship management (CRM), 74 management, 143 customer service, SLAs, 66

1532xix.fm Page 349 Monday, October 18, 2004 4:38 PM

development of application-development teams 349

pyramid model, 139–141 D responsibilities, 12 traffic, 274 data analyzers, 204 troubleshooting, 221–222 data collection, 108. See also baselining verifying, 61–63 data flow, application, 49 demarcation points, assigning, 128–130 data link connection identifier (DLCI), 88 dense wavelength-division multiplexing Data Link Switching (DLSw), 75 (DWDM), 88 data transfers, 290 dependency failures, 220 database administrator (DBA), 211 deployment database management system (DBMS), 160 application-deployment teams, 13 data-collection method, 159 applications, 50, 57 DBA (database administrator), 211 baselining, 155 DBMS (database management system), 160 big bang, 119–120 DE (Discard Eligible) bits, 135 contingency plans, 121–124 debug commands, 202–203 domains, 112–113 defining infrastructure, 107–108 base transactions, 43, 230 NetFlow, 165 business-critical applications, 22–24 pilot, 113–115 demarcation points, 130 policy management architecture, 109–112 location of metric collection points, 246–248 predictive modeling, 117–119 metrics, 153–159 profiling, 43–46 routes, 177 QoS resource allocation, 108–109 SLAs, 246–253 reporting, 124 transactions, 231, 248 staged, 119–120 delay TCO, 26 calculations, 182 teams, 13 CNS reporting, 213 verification, 115–117 distance, 191 VoIP, 260 introduction of, 27 WLANs, 281 measurement of, 108 deposit systems, 230 ping, 51 DES (discrete-event simulation), 50, 181 processing, 191 design protocols, 191 baselining, 48 queuing, 191 CLI, 109 SLAs, 65 management, 143 transmission, 190 metrics, 250 deleting leading/trailing data, 236 QoS, 109, 297 delivery RSPAN, 164 applications, 11 SPAN, 163 business objectives, 132–134 storage-area networks, 269–276 management, 141–151 TCO, 25–27 monitoring thresholds, 49 collecting metrics, 159–165 validation, 180 defining metrics, 153–159 WLANs, 279 identifying cycles, 167–168 development of application-development operating multiple systems, 151–153 teams, 13 reporting mechanisms, 165–167

1532xix.fm Page 350 Monday, October 18, 2004 4:38 PM

350 devices

devices DSSS (direct sequence spread spectrum), 276 capture points, 55 DWDM (dense wavelength-division classification, 112 multiplexing), 88 configuration, 109 DWRED (distributed weighted random early core, 113 detection), 98 CPE, 72, 295 dynamic environments, proactive planning, edge, 112 174–175 NAS, 271 Dynamic Host Configuration Protocol networks, 81 (DHCP), 84 PEPs, 110 storage-area networks, 269–276 DHCP (Dynamic Host Configuration Protocol), 84 E diagnostics, 142. See also troubleshooting CLI, 199–204 edge devices, QoS, 112 GUI, 204–206 edge nodes, 135 proactive troubleshooting, 175–178 efficiency, OSS, 141 systems, 77 EIGRP (Enhanced Interior Gateway Routing tools, 77 Protocol), 83 validating changes, 178–180 element management system (EMS), 145 differentiated services (DiffServ), 88 embedded agents, 72 DiffServ, 101–102 EMS (element management system), 145 DiffServ codepoint (DSCP), 101 emulation, traffic, 287 direct costs, TCO, 26 enabling direct sequence spread spectrum (DSSS), 276 debugging, 203 directory systems communication, 110 technologies, 180–181 disaster recovery. See also troubleshooting endpoints, TE, 89 planning, 123 end-to-end delivery systems tools, 121 enabling technologies, 180–181 Discard Eligible (DE) bits, 135 planning, 188–194 discovery, protocols, 158 end-to-end management, 188 discrete-event simulation (DES), 50, 181 end-to-end models, 284–297 distance delay, 191 end-to-end networks, 134–137 distributed weighted random early detection end-to-end QoS, 291 (DWRED), 98 end-to-end systems distribution, transactions, 232 APM, 7 division of end-to-end systems, 11 division of, 11 DLCI (data link connection identifier), 88 end-users, 26. See also users DLSw (Data Link Switching), 75 engineering, networks, 13 documentation, 25–27 Enhanced Interior Gateway Routing Protocol domain policies, 112–113 (EIGRP), 83 downstream alarm suppression, 142 enterprise APMs, 9. See also APMs downtime, 26 enterprise networks, 293 dropping, 136 enterprise operational support system, 141–151 frames, 251 environments schemes, 91 lab, 54 traffic, 135 production, 54 DSCP (DiffServ codepoint), 101 testing, 53

1532xix.fm Page 351 Monday, October 18, 2004 4:38 PM

GET statements 351

virtual, 50, 117 first-level help desks, 12 VoIP, 259–265 flat files, 160 equivalence, 183 flow ERC (error recovery code), 271 application data, 49 errors, 161. See also troubleshooting conversational paths, 247 essential applications. See also applications identifying, 81 defining, 22–24 profiling, 38 priority, 29–30 SANs, 273 requirements, 27–29 transactions, 241 TCO, 25–27 flow-based WFQ, 94 technical alignment, 30–32 flow-based WRED, 98 Ethernet collisions, 161 flowcharts etherStatsTable (polling), 160 pilot deployments, 114 ETSI (European Telecommunications Standards SLAs, 223 Institute), 315 troubleshooting, 178 evaluation, TCO, 26 validation, 179 events, 142 focus, key areas of business, 23 excess burst (Be), 135 forecasting resources, 10 expenses, TCO, 25–27 forwarding, 90, 135 explicit traffic, 118 forwarding equivalence class (FEC), 87 extensions, TE, 90 Frame Relay, 135 external probes, 72 frame-relay ip rtp header-compression command, 263 frames F average bytes per, 241 dropped/lost, 251 failures. See also troubleshooting FC, 271 analysis, 181 latency-bounded applications, 40 TCO, 25 threads, 40 fault windows, 254 fault-conditions, 219–220 framework systems, 198. See also monitoring management, 143 frequency of transactions, 24, 232 processing, 211 frequency hopping spread spectrum (FHSS), 276 solutions, 175. See also troubleshooting functions. See also commands FC (Fibre Channel), 269 business transactions, 229–246, 253–259 FCP (Fibre Channel Protocol), 271 investigating, 24 FDDI (Fiber Distributed Data Interface), 149 management systems, 143 FEC (forwarding equivalence class), 87 verification, 116 FHSS (frequency hopping spread spectrum), 276 future equipment costs, 26 Fiber Distributed Data Interface (FDDI), 149 Fibre Channel (FC), 269 Fibre Channel Protocol (FCP), 271 FIFO (first in, first out), 92–93 G files flat, 160 get command, 245 NFS, 271 get process (SNMP), 49 trace, 237–246 GET statements, 257

1532xix.fm Page 352 Monday, October 18, 2004 4:38 PM

352 GFR (guaranteed frame rate)

GFR (guaranteed frame rate), 294 information technology (IT) goals management, 211–215 lack of in methodologies, 31 organizational charts, 12 performance, 31–32 infrastructure setting, 28 baselining, 107–108 guaranteed frame rate (GFR), 294 planning, 29 GUI (graphical user interface), 152, 204–206 scaling, 285 VoIP, 260 installation TCO, 25–27 H WLANs, 281 interaction handing over problems, 289 applications, 217 hardware reduction of, 60 baselining, 48 video/voice, 290 external probes, 72 interconnection, SANs, 269 HBA (host bus adapter), 271 inter-exchange carrier (IXC), 291 help desks, 12 interfaces hierarchies, reporting, 212 APIs, 190 high utilization, 159 CLI, 109, 147 hooks, management of, 190 FDDI, 149 host bus adapter (HBA), 271 GUI, 152, 204–206 hub-and-spoke architecture, 135 IP tunnel, 90 hybrid simulation, 181 interior gateway protocols (IGPs), 90 International Organization for Standardization (ISO), 143 International Telecommunication Union (ITU), 315 I Internet Control Message Protocol (ICMP), 158, 190 Internet Engineering Task Force (IETF), 86, 303, 315 ICMP (Internet Control Message Protocol), Internet service providers (ISPs), 89, 291, 294 158, 190 Intserv (integrated services), 98–99 idle time, TCO, 26 intuition, 179 IETF (Internet Engineering Task Force), IP (Internet Protocol) tunnel interfaces, 90 86, 303, 315 ip rtp compression-connections command, 264 IGPs (interior gateway protocols), 90 ip rtp header-compression command, 263 ILEC (incumbent local exchange ISO (International Organization for carrier), 291 Standardization), 143 impact of business-critical applications, 22 isolation impact of new applications, 52 APM, 7–8 implementation. See also configuration events, 142 QoS, 96, 259–265 performance, 216. See also performance triggers, 13–14 ISPs (Internet service providers), 89, 291, 294 increasing IT (information technology) capacity, 61 management, 211–215 performance, 258. See also performance organizational charts, 12 incumbent local exchange carrier (ILEC), 291 ITU (International Telecommunication Union), 315 inefficient applications, 220 IXC (inter-exchange carrier), 291

1532xix.fm Page 353 Monday, October 18, 2004 4:38 PM

management 353

long-term monitoring, 218 J–L long-term triggers, 15 loss jitter priority, 29 priority, 29 QoS parameters, 108 profiling, 36 SLAs, 65 QoS parameters, 108 transaction captures, 236 SLAs, 65 lost frames, 251 VoIP, 261 low-latency queuing (LLQ), 95, 262 LSP (label switched path), 87 kbps (kilobits per second), 36 LSR (label switch router), 87 key areas of focus, 23, 29 key data collection points, 70 kilobits per second (kbps), 36 M KPIs (key performance indicators), 62, 76 MAC (Media Access Control), 149 lab environments, 54 maintenance. See also troubleshooting label switch router (LSR), 87 MTTR, 65 label switched path (LSP), 87 TCO, 25–27 label switching, 83–90. See also MPLS management latency accounting, 143 jitter, 36 APM, 5, 132–134 priority, 29 benefits of, 16 profiling, 35 end-to-end systems, 7 QoS parameters, 108 isolation, 7–8 traces, 218 lifecycles, 9 transactions, 236, 251 monitoring SLAs, 9 latency-bounded applications, 40, 58–59 overview of, 6 LDAP (Lightweight Directory Access Protocol), 110 sharing information, 8 leading data, deleting, 236 BSM, 133 leaky buckets, 96 business, 150–152 levels, baselining, 173–174 configuration, 143 LFNs (long fat networks), 275 CRM systems, 74 lifecycles, APM, 9 EMS, 145 Lightweight Directory Access Protocol (LDAP), 110 end-to-end models, 188, 284–297 limitations of responsibilities/skills, 130–131 events, 142 links fault, 143 DLSw, 75 hooks, 190 management, 90 IT organizational, 211–215 RF, 280 links, 90 link-state IGPs, 90 operations, 150 LLQ (low-latency queuing), 95, 262 oversight, 26 load thresholds, 250 performance, 144, 154 loading, 23–24 policies, 109–112 locations proactive application, 175 capture points, 234 pyramid model, 140 metric collection points, 246–248 QPM, 109, 204, 297 long fat networks (LFNs), 275 RF links, 280

1532xix.fm Page 354 Monday, October 18, 2004 4:38 PM

354 management

security, 145 messages, debug, 203 service, 150 metering, 136 SLAs, 67–68 methodologies SLM, 62 business and technical, 30–32 system, 13, 141–151 profiling, 43–46 TCO, 25 troubleshooting, 206–211 tools, 79 methods configuration, 83–84 collection, 72 DiffServ, 101–102 data-collection, 159 FIFO, 92–93 measurements, 72 IntServ, 98–99 of deployment, 120 MPLS, 85–88 metrics. See also requirements network resource allocation, 80–81 average bytes per frame, 241 non-work-conserving queuing, balancing, 154 96–98 collecting, 159–165 provisioning, 83–84 collection points, 246–248 queuing, 91–92 defining, 153–159 RSVP, 99–101 monitoring, 62 TE, 89–90 profiling, 61 traffic-handling mechanisms, 81–83 SLAs, 246–253 work-conserving queuing, 93–96 Cisco NetFlow, 73–74 traffic, 285–286 defining, 65–66 users, 187 delivery, 66–67 Management Information Bases (MIBs), management, 67–68 48, 118, 200 monitoring, 76–77 Management Interface Bases (MIBs), 147 profiling, 68–76 managers (SNMP), 147 tracking, 249–253 mapping MIBs (Management Information Bases), conversational, 240 48, 118, 200 priorities, 185 MIBs (Management Interface Bases), 147 services, 74–76 milliseconds (ms), 108 markers, 135 mirroring ports, 163 marking, 102, 135–136 mission-critical applications, 291 matching demarcation points, 130 mission-critical activities, 7 maximum transmission unit (MTU), 263 models mean time to repair (MTTR), 65 analytical, 51 measurements, 159 business management, 152 availability, 157 delivery responsibilities, 12 bandwidth, 36 DES, 51 capacity planning, 184 end-to-end, 284–297 CPE/core networks, 72 predictive modeling, 117–119 delay, 108 pyramid, 139–141 external probes, 72 scenario, 183 latency, 40 simulation, 45, 52 methods, 72 virtual, 178 NetFlow, 165 virtual-environment methods, 50 Media Access Control (MAC), 149 modification merging statistics, 237–246 baselining, 48

1532xix.fm Page 355 Monday, October 18, 2004 4:38 PM

networks 355

CLI, 109 management, 143 N metrics, 250 NAS (network-attached storage), 271–272 QoS, 109, 297 NBAR (network-based application recognition), 81 RSPAN, 164 NE (network element), 145 SLAs, 259 near-real-time network monitoring capabilities, 161 SPAN, 163 NetFlow, 73–74, 161, 165 storage-area networks, 269–276 network element (NE), 145 TCO, 25–27 Network File System (NFS), 149, 271 thresholds, 49 network interface cards (NICs), 76, 281 validation, 180 network management systems (NMS), 49, 145 monitoring Network Node Manager (NNM), 49 agents, 192 network operations center (NOC), 154 APM, 127, 132–134 network service providers (NSPs), 291 applications, 198, 205 networkability, 15 average bytes per frame, 241 network-attached storage (NAS), 271 capacity planning, 184 network-based application recognition (NBAR), 81 delivery networks classifying management, 141–151 analyzers, 38, 234 collecting metrics, 159–165 BSM, 133 defining metrics, 153–159 CNS delay reporting, 213 identifying cycles, 167–168 devices, 81 multiple systems, 151–153 engineering, 13 pyramid model, 139–141 enterprise, 293 reporting mechanisms, 165–167 forecasting, 10 demarcation points, 128–130 latency, 35 end-to-end delivery systems, 192 monitoring, 198 Ethernet collisions, 161 operations, 13 long-term, 218 pipes. 159 metrics, 61–62, 249–253 planning, 13 networks, 198 resource allocation, 80–81 period of, 62 security, 13 polling, 162 SLAs ports, 163 Cisco NetFlow, 73–74 predictive modeling, 117–119 defining, 65–66 real-time troubleshooting, 66 delivery, 66–67 resource-consumption data, 10 management, 67–68 SLAs, 9, 66, 76–77 monitoring, 76–77 toolsets, 217 profiling, 74–76 MPLS (Multiprotocol Label Switching), 83–88 requirements, 68–73 ms (milliseconds), 108 SP, 262 MTTR (mean time to repair), 65 statistics, 237–246 MTU (maximum transmission unit), 263 TCO, 25–27 multicast traffic (etherStatsMulticastPkts), 161 transmission paths, 294 multimedia applications, 217 unified views of, 176 multiple systems, operating, 151–153 VPNs, 65 Multiprotocol Label Switching (MPLS), 83–88 WLANs, 276–284

1532xix.fm Page 356 Monday, October 18, 2004 4:38 PM

356 NFS (Network File System)

NFS (Network File System), 149, 271 NICs (network interface cards), 76, 281 P NMS (network management systems), 49, 145 packet-by-packet traffic, 177 NNM (Network Mode Manager), 49 packets NOC (network operations center), 154 BPDU, 204 nodes, 135 flow, 38 non-work–conserving queuing mechanisms, latency, 35 96–98 loss notification, single-point-of-failure, 142 SLAs, 65 NSPs (network service providers), 291 transaction captures, 236 number of metrics, selecting, 154 marking, 102 numbers, ports, 243 webtraffic, 265 parallel fault processing, 212 passive agents, 192 O passive probes, 72 paths object identifier (OID), 48, 147 capture points, 234 objectives, loading, 23 conversation, 157 objects, polling, 49 conversational flows, 247 OBJECT-TYPE constructs, 148 LSP, 87 OID (object identifier), 48, 147 TE, 89–90 Open Shortest Path First (OSPF), 83 transmission, 294 operating costs, TCO, 25 patterns, traffic, 107–108 operating multiple systems, 151–153 PDP (policy decision point), 109 operational objectives, 174–175 PDUs (protocol data units), 147 operational support system (OSS), 141–151 peer support, 26 operational systems, 77 penalties, SLAs, 187 operations management, 150. See also management PEPs (policy enforcement points), 109 OPNET performance. See also optimization ACE, 234 APM, 5, 127 IT Guru, 216 benefits of, 16 optimization end-to-end systems, 7 architecture, 256 isolation, 7–8 bandwidth-bounded applications, 59–61 lifecycles, 9 delivery, 221–222 meeting service needs, 132–134 end-to-end models, 286 monitoring SLAs, 9 latency-bounded applications, 58–59 overview of, 6 SAN traffic delivery, 274 sharing information, 8 servers, 255 baselining, 173–174 TE, 89 demarcation points, 128–130 transactions, 253–259 dynamic environments, 174–175 windows, 254 end-to-end models, 284–297 organizational charts, IT, 12 goals, 31–32 OSPF (Open Shortest Path First), 83 increasing, 258 OSS (operational support system), 141–151 KPI, 76 overloading clients/servers, 220 management, 144, 154 oversight, management, 26 polling, 163

1532xix.fm Page 357 Monday, October 18, 2004 4:38 PM

processes 357

SLAs, 246–253 SNMP, 161 Cisco NetFlow, 73–74 threshold, 162 defining, 65–66 portfolio transactions, 230 delivery, 66–67 ports management, 67–68 allocation, 243 monitoring, 76–77 critical switched connections, 71 profiling, 74–76 mirroring, 163 requirements, 68–73 monitoring, 163 troubleshooting, 142, 180, 216, 221 POST statements, 257 permanent virtual circuit (PVC), 261 PQ (priority queuing), 81, 94, 261 phases, capacity planning, 185 PQCBWFQ (priority queue class-based weighted PHB (per-hop behavior), 101 fair queuing), 95 physical connectivity, transactions, 232 predeployment testing, 116. See also deployment pilot deployments, 113–115 predictability, SLAs, 65 ping predicted access points, transactions, 232 delay, 51 predictive modeling, 117–119 response times, 158 predictive performance analysis, 15 planning, 13 priorities, mapping, 185 capacity, 66, 174, 183–186 prioritization, 184 deployment, 121–124 business-critical applications, 29–30 disaster recovery plans, 123 CLP, 135 dynamic environments, 174–175 priority queue class-based weighted fair queuing end-to-end delivery, 188–194 (PQCBWFQ), 95 infrastructure, 29 priority queuing (PQ), 81, 94, 261 networks, 13 priority schemes, 23 performance, 173–174 proactive fault processing, 212 pilot deployments, 115 proactive planning proactive troubleshooting, 175–178 capacity planning, 183–186 enabling technologies, 180–181 dynamic environments, 174–175 validating changes, 178–180 end-to-end delivery, 188–194 scenario modeling, 181–183 performance, 173–174 SLMs, 186–188 scenario modeling, 181–183 systems, 13 SLMs, 186–188 policies, 24 troubleshooting, 175–178 architecture management, 109–112 enabling technologies, 180–181 domains, 112–113 validating changes, 178–180 QoS, 109, 297 probes suppression, 142 application-aware, 158 policing, 135 capture, 234 policy decision point (PDP), 109 external, 72 policy enforcement points (PEPs), 109 passive, 72 polling, 146, 161 profiling, 39 data, 160 software, 72 monitor, 162 processes OIDs, 48 capture, 234 performance, 163 OSS, 141 records, 49 timing of transactions, 24

1532xix.fm Page 358 Monday, October 18, 2004 4:38 PM

358 processing

processing, 190 configuration, 83–84 delay, 191 core devices, 113 faults, 211 devices, 81 time, 252 DiffServ, 101–102 production environments, 54 edge devices, 112 productivity, TCO, 25 end-to-end, 291 profiling, 24, 35 FIFO, 92–93 agents, 39 implementation, 96 analyzers, 38 IntServ, 98–99 applications, 29, 37, 40 network resource allocation, 80 bandwidth, 36 policies, 297 conversation, 38 provisioning, 83–84 flow, 38 queuing, 91–92 jitter, 36 resource allocation, 108–109 latency, 35 RFCs, 303–312 methodologies, 43–46 RSVP, 99–101 metrics, 61 services, 294 probes, 39 standards SAA, 39 ETSI, 316 SLAs, 74–76 IETF, 315 transactions, 229–246 ITU, 315–316 profits, TCO, 27 traffic-handling mechanisms, 81–83 projecting future resource needs, 184 troubleshooting tools, 198 propagation, 190 application monitoring, 198 protocol data units (PDUs), 147 network monitoring, 198 protocols, 190 VoIPs, 259–265 analyzers, 38 WAN, 295 baselining, 107–108 QPM (QoS Policy Manager), 109, 204, 297 delay, 191 quality, APM, 6 discovery, 158 quality of service. See QoS effect of, 51 quantity of metrics, selecting, 154 QoS, 134–137 queries routing, 83 amount of data transferred, 245 SNMP, 49 analysis, 257–258 TCP, 37 repeating, 244 provisioning network devices, 81 queuing, 81, 91–92, 190 PSTN (Public Switched Telephone algorithms, 82 Network), 259 business transactions, 230 purchasing, TCO, 25 delay, 191 PVC (permanent virtual circuit), 261 FIFO, 92 pyramid model, monitoring delivery, 139–141 simulators, 182 types of, 81 Q R QoS (quality of service), 35, 84 802.11 networks, 276 radio frequency (RF), 277, 280 allocation, 297 RAID (redundant array of inexpensive disks), 271 assigning, 134–137 random early detection (RED), 81, 98

1532xix.fm Page 359 Monday, October 18, 2004 4:38 PM

scaling 359

RCA (root-cause analysis), 142, 150 response times, 51, 157 reallocation, 184 APM, 6 Real-Time Transport Protocol header (cRTP), 263 business-critical application requirements, 28 real-time troubleshooting, 66 SLAs records defining, 65–66 capture time, 236 delivery, 66–67 polling, 49 responsibilities transaction time, 239 delivery models, 12 UDP, 165 limitations of, 130–131 recovery results, profiling, 46 disaster, 121 retransmissions, TCP, 255–256 ERC, 271 revenue RED (random early detection), 81, 98 key areas of business focus, 23 reduction streams, 24 of data, 59 TCO, 25–27 of interactions, 60 reviewing of window sizes, 60 SLAs, 186–188, 222–224 redundancy, 157 traces, 237 redundant array of inexpensive disks RF (radio frequency), 277, 280 (RAID), 271 RFC (Request For Comment), 303–312 relocation of servers, 61 RMON (remote monitoring), 149 re-marking traffic, 135 rollouts, 115, 185 remote monitoring (RMON), 149 root-cause analysis (RCA), 142, 150 Remote Switched Port Analyzer (RSPAN), 163 round-trip time (RTT), 275 repairability of MTTR, 65 routers, LSR, 87 repeating queries, 244 routes, defining, 177 reporting, 165–167 routing CNS delay, 213 constraint-based, 90 deployment, 124 protocols, 83 hierarchies, 212 ns, 90 profiling, 46 RSVP (Resource Reservation Protocol), to control consoles, 192 90, 99–101 Request For Comment (RFC), 303–312 RTT (round-trip time), 275 requests, VC, 86 rules requirements capture, 236 business, 127 SLAs, 130 business-critical applications, 27–29 delivery, 221–222 end-to-end delivery systems, 188 S QoS, 108 SLAs, 68–73 SAA (Service Assurance Agent), 39, 72, 158, 199 Resource Reservation Protocol (RSVP), 90 SAA/RTR (Service Assurance Agent/Response resources Time Reporter), 205 allocation, 80–81 SACK (selective acknowledgement), 276 forecasting, 10 SANs (storage-area networks), 269–276 monitoring, 10 scaling QoS allocation, 108–109 infrastructure, 285 requirements, 231 TCP windows, 275

1532xix.fm Page 360 Monday, October 18, 2004 4:38 PM

360 scenario modeling

scenario modeling, 181–183 simulation schemes DES, 181 dropping, 91 hybrid, 181 queuing, 91 models, 45, 52 security simulators, calculations, 182 management, 145 single-point-of-failure notification, 142 networks, 13 single-transaction monetary functions, 231 selection of deployments, 119–120 site surveys, WLANs, 277–279, 281 selective acknowledgment (SACK), 276 sizes, windows, 60, 254 serialization, 190 skills, limitations of, 130–131 server message block (SMB), 76 SLAs (service level agreements), 9, 102,130 servers, 269–276 Cisco NetFlow, 73–74 CNS delay reporting, 213 compliance analysis, 180 conversational maps, 240 defining, 65–66, 246–253 optimization, 255 delivery, 66–67 overloading, 220 management, 67–68 processing time, 252 monitoring, 76–77 relocating, 61 penalties, 187 Service Assurance Agent (SAA), 39, 72, 158, 199 profiling, 74–76 Service Assurance Agent/Response Time Reporter requirements, 68–73 (SAA/RTR), 205 reviewing, 222–224 service level agreements. See SLAs SLM (service level management), 9, 62, 186–188 service level management, 9, 62, 186–188 slow delivery, TCO, 25 service provider (SP) networks, 262 SMB (server message block), 76 services, 81 Sniffer, 234 APM, 6 SNMP (Simple Network Management BSM, 133 Protocol), 147 customer, 66 agents, 148 IntServ, 98–99 data-collection methods, 160 management, 150 get process, 49 mapping, 74–76 MIB, 48 meeting needs of, 132–134 polling, 161 NetFlow, 165 snmpwalk tool, 48–49 QoS, 294 soft costs, 26 troubleshooting, 222–224 software validation, 66 baselining, 48 sessions, RSPAN, 164 probes, 72 setsockopt, 254 SP (service provider) networks, 262 shaping traffic, 135, 275–276 SPAN (Switched Port Analyzer), 163 sharing information, APM, 8 specialized support, 12 show commands, 201 SPF (shortest path first), 90 show processes cpu command, 204 staged deployments, 119–120 shortest path first (SPF), 90 standards, QoS, 315–316 signaled mechanisms, 83 statements Simple Network Management Protocol GET, 257 (SNMP), 147 inquiries, 230 simple transactions, 230 POST, 257

1532xix.fm Page 361 Monday, October 18, 2004 4:38 PM

tools 361

statistics retransmissions, 255–256 error, 161 SAN, 274 merging, 237–246 turns, 37, 243 traffic, 161 windows types of, 51 scaling, 275 status, applications, 43 sizing, 254 storage-area networks (SANs), 269–276 TDM (time-division multiplexing), 95 strategic connection points, demarcation points, 130 TE (traffic engineering), 88–90 strategies, APM, 10–11 teams streaming video/voice, 290 application-deployment, 13 streams, revenue, 24 application-development, 13 stress, verifying functions under, 116 technical alignments, 30–32 sub-branches, transactions, 232 technical baselines, 121 support technical criteria category (SLAs), 69–70 demarcation points, 128 technical requirements, 27 future services, 185 technologies, enabling, 180–181 peer, 26 telecommunication costs, 26 specialized, 12 Telnet, 147, 159 TCO, 25 testing. See also troubleshooting; verification suppression of events, 142 environments, 53 Switched Port Analyzer (SPAN), 163 predeployment, 116 switched port connections, 71 threads, applications, 40 switches, LSR, 87 thresholds, 142 switching, DLSW, 75 analyzing, 49 synthetic transactions, 75 capacity, 185 syslog protocol, 149 load, 250 systems, 13 metrics, 250 administration, 13 polling, 162 BSM, 133 time DBMS, 160 processing, 252 design, 25 RTT, 275 EMS, 145 transactions, 239 management, 141–151 time-division multiplexing (TDM), 95 multiple, 151–153 time-sensitive conditions, thresholds, 142 NFS, 149 timing, 24 NMS, 145 token buckets, 96 polling, 146 tools collection, 249 diagnostic, 199 CLI, 199–204 T GUI, 204–206 disaster recovery, 121 tail drop, 82, 97, 135 management, 79 targets, transactions, 248 configuration, 83–84 tasks, 38, 217 DiffServ, 101–102 TCO (total cost of ownership), 25–27 FIFO, 92–93 TCP (Transmission Control Protocol) IntServ, 98–99 ACKs, 254 MPLS, 85–88

1532xix.fm Page 362 Monday, October 18, 2004 4:38 PM

362 tools

network resource allocation, 80–81 traffic-conditioning implementations, 134 non-work-conserving queuing, 96–98 traffic-handling mechanisms, 81–83 provisioning, 83–84 traffic-shaping mechanisms, 96 queuing, 91–92 trailing data, deleting, 236 RSVP, 99–101 trances, cleaning, 236 TE, 89–90 transactions, 38, 217 traffic-handling mechanisms, 81–83 base, 43 work-conserving queuing, 93–96 capturing, 236 pyramid model, 139–141 defining, 27 snmpwalk, 48–49 flow, 241 troubleshooting, 198 optimizing, 253–259 application monitoring, 198 profiling, 229–246 network monitoring, 198 synthetic, 75 toolsets, monitoring, 217 targets, 248 top-down (pushed) mechanisms, 83 timing, 24 topologies, building, 118 transfers ToS (Type of Service), 101 amount of data, 245 total cost of ownership (TCO), 25–27 data, 290 total traffic (etherStatsOctets), 161 transmissions traces delay, 190 capturing, 218–219, 234–235 paths, 294 files, 237–246 traps, 147 reviewing, 237 triggers tracking implementation, 13 metrics, 249–253 transactions, 250 SLAs, 246–253 troubleshooting, 142 traffic, 118 delivery, 221–222 capturing, 218 dependency failures, 220 Cisco IO traffic-conditioning mechanisms, 136 diagnostic tools, 199 defining, 177 CLI, 199–204 delivery, 274 GUI, 204–206 dropping, 135 disaster recovery, 121 emulation, 287 end-to-end models, 284–297 end-to-end models, 284–297 fault-conditions, 219–220 jitter, 261 flowcharts, 178 network resource allocation, 80 IT organizational, 211–215 packet-by-packet, 177 jitter, 261 patterns, 107–108 methodologies, 206–211 QoS, 134–137 MTTR, 65 real-time troubleshooting, 66 performance, 180, 216 re-marking, 135 proactive, 175–180 RSPAN, 164 real-time, 66 shaping, 135, 275–276 services, 222–224 simulation models, 45 tools, 198 simulators, 182 transactions, 238 statistics, 161 triggers, 14 voice (WLANs), 282 WLANs, 281 webtraffic packets, 265 tuning, 289. See also optimization traffic engineering (TE), 88–90 tunneling, IP interfaces, 90

1532xix.fm Page 363 Monday, October 18, 2004 4:38 PM

WRED (weighted random early detection) 363

turns virtual circuit (VC) requests, 86 application, 37 virtual environments TCP, 243 creating, 117 Type of Service (ToS), 101 scenario modeling, 182 types troubleshooting, 176 of queuing, 81 validating, 179 of statistics, 51 virtual models, creating, 178 virtual private networks (VPNs), 65, 86 virtual-environment modeling methods, 50 U–V VLANs (virtual LANs) key data collection points, 71 UBR (unspecified bit rate), 294 RSPAN, 164 UDP (User Datagram Protocol), 165 voice unified views of networks, 176 interactive, 290 UNIX, snmpwalk tool, 48–49 streaming, 290 unspecified bit rate (UBR), 294 WLANs, 282 updates, SLAs, 259 VoIP (Voice over IP), 96, 259–265, 282 upgrades, 221 VPNs (virtual private networks), 65, 86 User Datagram Protocol (UDP), 165 users management, 187 W–Z TCO, 25 transactions, 232 WAN (wide-area network) utilization, 159 jitter, 261 QoS, 295 validation SP clouds, 262 configuration, 180 web-based applications, 291 services, 66 webtraffic packets, 265 values, business, 231 weighted fair queuing (WFQ), 81 VC (virtual circuit) requests, 86 weighted random early detection (WRED), 81, 98, vendor specifications, 179 135, 276 variable bit rate (non-real time), 294 WFQ (weighted fair queuing), 81 variable bit rate (real time) (VBR), 294 wide-area network. See WAN verification windowing, TCP, 254 delivery, 61–63 windows deployment, 115–117 sizing, 60, 254 performance increase, 258 TCP, 275 video wireless LANs, 276–284 interactive, 290 wires, reduction of data on, 59 streaming, 290 WLANs (wireless LANs), 276–284 videoconferencing, 108 work-conserving queuing, 93–96 viewing workloads, applications, 10 average bytes per frame, 241 WRED (weighted random early detection), 81, 98, profiling results, 46 135, 276