click below
click below
Normal Size Small Size show me how
csc chp4 flashcards
topic = what's inside a router
Question | Answer |
---|---|
how fast does a routing/management control plane work | operates in millisecond timeframe |
how fast dpes forwarding data plane work | operates in nanosecond timeframe |
what is one function of an input port function | decentralizing switching |
what does decentralized switching do | using header field values, looks up output port using forwarding table in input port memory |
what is the goal of decentralized switching | complete input port processing at line speed |
what/how does input port queueing work | if datagrams arrive faster than forwarding rate into switch fabric |
what is destination-based forwarding | forward based only on destination IP address (traditional) |
what is generalized forwarding | forward based on any set of header field values |
destination-based forwarding example | slide 17 |
how does longest prefix matching work | when looking for forwarding table entry for given destination address, use longest address prefix that matches destination address |
longest prefix matching example | slide 18-21 |
what does longest prefix matching often do | performed using ternary content addressable memories (TCAMs) |
what is content addressable | present address to TCAM = retrieve address in one clock cycle, regardless of table size |
what is a cisco catalyst | 1M routing table entries in TCAM |
what is the definition of switching fabrics | transfer packet from input lint to appropriate output link |
what is switching rate | rate at which packets can transfer from inputs to outputs (often measured as multiple of input/output line rate / N inputs = switching rate N times line rate desirable) |
what are the three major types of switching fabrics | memory / bus / interconnection network |
what's one method to switch via memory | first generation routers |
how do first generation routers work | traditional computers with switching under direct control of cpu / packet copied to system's memory / speed limited by memory bandwidth (2 bus crossing per datagram) |
how does switching via bus work | datagram from input port memory to output port memory via shared bus |
what is a bus contention | switching speed limited by bus bandwidth |
what is sufficient speed for access routers for switching via bus | 32 gbps bus, cisco 5600 |
how does switching via interconnection network work | crossbar, clos networks, other interconnection nets initially developed to connect processors in multiprocessor |
what is a multistage switch | nxn switch from multiple stages of smaller switches |
define exploiting parallelism | fragment datagram into fixed length cells on entry / switch cells through the fabric, reassemble datagram at exit |
what is scaling regarding switching via interconnection network | using multiple switching "planes" in parallel (helps speedup process) |
how does a cisco CRS router work | basic unit = 8 switching planes / each plane = 3-stage interconnection network / up to 100's tbps switching capacity |
what is input port queuing useful for | if switch fabric slower than input ports combined > queuing ay occur at input queues (queueing delay and loss due to input buffer overflow) |
define head-of-the-line blocking | queued datagram at front of queue prevents others in queue from moving forward |
how does buffering work for output port queuing | required when datagrams arrive from fabric faster than link transmission rate |
what is a drop policy | which datagrams to drop if no free buffers |
what is scheduling discipline | chooses among queued datagrams from transmission |
how can datagrams be lost | due to congestion or lack of buffers |
what does priority scheduling provide | network neutrality, gets best performance |
when does buffering take place in output port queuing | when arrival rate via switch exceeds output line speed |
why does queueing and loss occur | due to output port buffer overflow |
formula for buffering | slide 32 |
what can too much buffering produce | increase in delays (specifically home routers) |
how do long RTTs work | poor performance for realtime apps (sluggish TCP response) |
recall what delay-based congrestion control is | keep bottleneck link just full enough but no fuller |
what are the two methods for buffer management | drop and marketing |
what is meant by drop in buffer management | which packet to add/drop when buffers are full |
what is a tail drop in buffer management | drop arriving packet |
what is a priority in buffer management | drop/remove on priority basis |
what is marketing in buffer management | which packets to mark to signal congestion (ECN/RED) |
what is packet scheduling | deciding which packet to send next on link |
what are the different links in a packet | first come first served / priority / round robin / weighted fair queueing |
what is FCFS regarding packet scheduling | packets transmitted in order of arrival to output port (also know as FIFO) |
what is the purpose of scheduling policies | arriving traffic classified, queued by class (any header fields can be used for classification) / send packet from highest priority queue that has buffered packets (FCFS within priority class) |
how does round robin scheduling work | arriving traffic classified queued by class (any header fields can be used for classification) / server cyclically repeatedly scans class queues sending one complete packet from each class |
how does a weighted fair queuing work | each class has a weight and gets weighted amount of service in each cycle (min bandwidth guarantee) - generalized round robin |
what is network neutrality | how an ISP should share/allocate its resources using mechanisms like packet scheduling or buffer management |
what are social economic principles | protecting free speech and encouraging innovation and competition |
what does network neutrality enforce | rules and policies |
what does "no blocking" mean in network neutrality | shall not block lawful content, applications, services or non-harmful devices subject to reasonable network management |
what does "no throttling" mean in network neutrality | shall not impair or degrade lawful internet traffic on the basis of internet content, application or service or use of a non-harmful device |
what does "no paid prioritization" mean in network neutrality | shall not engage in paid prioritization |
END | next section from slide 42 |