01 05 Computer System Architecture — Transcript
Full transcript
- 0:08[Music]
- 0:21computer systems architecture
- 0:24so different operating systems for
- 0:26different kinds of computer environments
- 0:31so most systems use a single general
- 0:34purpose processor
- 0:36most systems have special purpose
- 0:38processor as well
- 0:39okay one main cpu which manages the
- 0:43computer and runs user apps
- 0:47other specialized processor this
- 0:49controllers gpus etc
- 0:51do not run user apps
- 0:54okay
- 0:55now for the multiprocessor systems
- 0:59it increased throughput
- 1:02okay faster execution but not 100
- 1:06percent linear speed up
- 1:09next would be the economy of scale
- 1:12peripherals disks
- 1:14memory shared among processors
- 1:18and increased reliability
- 1:20you've got something like
- 1:22failure of cpu
- 1:25allows
- 1:27our slow system
- 1:29don't crush it
- 1:30okay redundant processing provides
- 1:33system of checks and balances
- 1:35so graceful degradation
- 1:38or degradation
- 1:40of fault tolerance
- 1:42so there are two types
- 1:44okay so you've got the
- 1:47asymmetric multiprocessing
- 1:49wherein each processor is assigned
- 1:51specific
- 1:53okay
- 1:54and
- 1:55symmetric multiprocessing
- 1:58each processor performs all tasks
- 2:04now i have here an example of symmetric
- 2:06multi-processing architecture okay
- 2:09so you've got processor zero and
- 2:11processor one here
- 2:13sharing the
- 2:15main memory
- 2:16each cpu has their own register and
- 2:18cache
- 2:19but the main memory are shared
- 2:24how about on the dual core design
- 2:26so multi-chip and multi-core so systems
- 2:30containing all chips
- 2:32chassis containing multiple separate
- 2:34systems
- 2:35okay so something like we have one
- 2:37physical processor
- 2:39virtually divided or virtually having
- 2:42multiple cores
- 2:43in this case you've got dual core
- 2:46okay cpu core zero has their own
- 2:48registers and has their own cache
- 2:51and so with
- 2:53the core one but they share an l2 cache
- 2:57okay leading to the main memory
- 3:02for the non-uniform memory access system
- 3:05okay so one core
- 3:08could switch
- 3:09okay from
- 3:11one cpu to the next
- 3:13okay
- 3:14so something like okay so i have here
- 3:16memory zero and i have memory one here
- 3:19the cpu
- 3:21is accessing the memory
- 3:24okay
- 3:25in non-uniform fashion
- 3:28all right
- 3:31plastered system
- 3:33so independent systems with shared
- 3:36common storage
- 3:39and connected by a high-speed land
- 3:42working together
- 3:44okay
- 3:45so special considerations for access to
- 3:47shared storage are required
- 3:49we have this uh distributed lack
- 3:51management
- 3:54as our collaboration protocols
- 3:57so a good example of this is
- 4:00the implementation of the storage area
- 4:02network
- 4:03okay so it provides a high availability
- 4:06service
- 4:07which survives failures
- 4:10so there are two types for clustered
- 4:12systems you've got symmetric and
- 4:14asymmetric clustering
- 4:17so asymmetric has one machine
- 4:20in hot standby mode
- 4:22will symmetric has multiple nodes
- 4:25running applications monitoring each
- 4:27other okay so some clusters are for high
- 4:30speed
- 4:31performance computing or hpc
- 4:34applications must be written to use
- 4:35parallelization
- 4:38okay
- 4:39some have distributed lag manager or dlm
- 4:43okay the distributed log management
- 4:45to avoid conflicting operations
- 4:50so this is an example of clustered
- 4:52systems
- 4:53we have these computers sharing a
- 4:56storage
- 4:57area
- 4:58and this storage area network here
- 5:01our network of hard disks
- 5:03okay so which are being shared
- 5:06by
- 5:07the computer on the given infrastructure
- 5:13next pc motherboard so consider the
- 5:16desktop pc motherboard with a processor
- 5:18socket shown below so you've got your
- 5:20processor socket you've got the memory
- 5:24slots there
- 5:25and then these are the buses
- 5:28so this board is a fully functioning
- 5:30computer
- 5:31once it's slots are populated something
- 5:34like if you're going to install your ram
- 5:36here
- 5:37you've got for instance
- 5:39the pci but then we already have our
- 5:42memory or the
- 5:44display interface here just like in the
- 5:46keyboard
- 5:48okay
- 5:49so even the lowest cost general purpose
- 5:51cpu contains multiple cores
- 5:54some motherboards contain multiple
- 5:56processor sockets more advanced
- 5:58computers allow more than one system
- 6:01creating a pneuma system
- 6:04okay so basically
- 6:06if you have nothing to
- 6:08install here like you don't want to use
- 6:10your display adapter and you use this
- 6:13instead
- 6:14then that would make it operational
- 6:17okay
- 6:26[Music]
- 6:34you
About this transcript
This page contains the full transcript of 01 05 Computer System Architecture by Santelmo, generated from the public captions YouTube serves with the video. The transcript has 624 words across 153 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.
What you can do with it
Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to translate the transcript into 100+ languages.
Free YouTube transcript tool
YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, translation to 100+ languages, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.