Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 4

Basics of Operating Systems (IT2019-2)

GIANG LE e1800927
Assignment V:
1. What is direct memory access (DMA)? Why is it needed? What is UDMA?
 DMA is a feature that allows external devices send registers to the main memory
without any CPU intervention. When the transfer is done, it sends a single interrupt to
CPU
 DMA is needed to avoid sending too many interrupts to CPU, that allows CPU performs
other works, improving performance
 UDMA (Ultra DMA) is the fastest protocol used to transfer data through the ATA
controller, which is an interface that connects hard drives, CD-ROM drives and other
drives

2. What is the difference between Switch and Bus technologies on a Computer system?
 In Bus technology, the path to different components of the network are shared -> slow
 In Switch technology, there are many concurrent paths to different components, these
paths are point to point -> faster

3. What are differences between single-processor and multiprocessor computer systems?


 Single processor system contains only one processor while multiprocessor systems may
contain two or more processors.
 Single processor systems use different controllers for completing special tasks such as
DMA (Direct Memory Access) Controller. On the other hand, multiprocessor systems
have many processors that can perform different tasks.
 Single processor systems can be more expensive than multiprocessor systems.
 It is easier to design a single processor system as compared to a multiprocessor system.
 Throughput of a multiprocessor system is more than a single processor system.
 Single processor systems are less reliable than multiprocessor systems because if the
processor fails for some reason then system cannot work.
 Most modern personal computers are single processor systems while multiprocessors
are used in niche systems only.

4. What do microprocessors do in a computer system?


The microprocessor is the central unit of a computer system that performs arithmetic and
logic operations, which generally include adding, subtracting, transferring numbers from
one area to another, and comparing two numbers. It incorporates the functions of a
central processing unit on a single (or more) integrated circuit (IC) of MOSFET construction.
The microprocessor is a multipurpose, clock-driven, register-based, digital integrated

1
circuit that accepts binary data as input, processes it according to instructions stored in its
memory, and provides results (also in binary form) as output.

5. How does CPU take care of disk scheduling?


A disk-controller microprocessor receives a sequence of requests from the main CPU and
implements its own disk queue and scheduling algorithm. This arrangement relieves the
main CPU of the overhead of disk scheduling.

6. What is graceful degradation? What is fault tolerance?


 Graceful degradation is the ability of a computer, machine, electronic system or
network to maintain limited functionality even when a large portion of it has been
destroyed or rendered inoperative. The purpose of graceful degradation is to prevent
catastrophic failure. (the system can work if there is any problem, even slower)
 Fault tolerance is the property that enables a system to continue operating properly in
the event of the failure of (or one or more faults within) some of its components. (the
other processors will continue the work if one has failure)

7. What is the requirement to have a fault tolerant system?


The basic characteristics of fault tolerance require:
 No single point of failure – If a system experiences a failure, it must continue to operate
without interruption during the repair process.
 Fault isolation to the failing component – When a failure occurs, the system must be
able to isolate the failure to the offending component.
 Fault containment to prevent propagation of the failure – Some failure mechanisms can
cause a system to fail by propagating the failure to the rest of the system. Firewalls or
other mechanisms that isolate a rogue transmitter or failing component to protect the
system are required.
 Availability of reversion modes

8. How does a HP NonStop system work?


 The HP NonStop Operating System uses a message-based architecture assisted by
ServerNet hardware features to connect from two to 16 independent processors within
each highperformance node. An instance of the NonStop Operating System resides in
every processor.
 The NonStop OS implements software fault tolerance, monitors and synchronizes all
processors, coordinates resource sharing, and provides access control for all ServerNet
connected devices. Each processor has its own memory and uses ServerNet technology
to access I/O devices, other processors, and other HP NonStop servers.

9. What is the difference between asymmetric and symmetric multiprocessing?


 In asymmetric multiprocessing, there is one boss processor instructs other processors
do it task, and each processor do a specific task (boss-worker relationship)
 In symmetric multiprocessing, all processors are peers, trying to solve one problem,
and share the same memory

10. What could be the drawback (disadvantages) of a SMP system? What is the solution?

2
 The Operating System and programs/applications running an SMP system need to be
compatible with it; otherwise the multiple processors are completely wasted.
 Cannot use floating-point based processors
 There are limits to SMP systems due to scalability limits of modern processors.
 The master CPU can become bottlenecked (a bottleneck being a single part of a system
that ends up completely crippling the system’s ability to function the way it’s intended
to).

Solution:
 I/O must be carefully controlled to ensure that the data reach the appropriate
processor. Also, since the CPUs are separate, one may be sitting idle while another is
overloaded, resulting in inefficiencies.
 These inefficiencies can be avoided if the processors share certain data structures. A
multiprocessor system of this form will allow processes and resources—such as
memory—to be shared dynamically among the various processors and can lower the
variance among the processors.

11. What is uniform memory access (UMA)? What is non-uniform memory Access (NUMA)?
 UMA is defined as the situation in which access to any RAM from any CPU takes the
same amount of time.
 NUMA is when some parts of memory take longer to access than other parts

12. What are the advantages of a multicore computer system?


 More efficiency, because it can get more work done in the same unit of time
 Lower cost, because it can share peripherals, mass storage and power supplies
 More reliable, because there are alternative cores can do the work if there is any
failure in one core

13. What is the difference between blade and traditional multiprocessor systems?
The different between blade and traditional multiprocessor systems is that each blade-
processor board boots independently and runs its own operating-system.

14. What is the difference between a clustered system and a multiprocessor system?
 In clustered system, individual systems are tied together and multiprocessor system is
only single system
 LAN networking is used in clustered system, in multiprocessor system LAN is not
required
 The aim is high availability in clustered system whereas in multiprocessing it is time
saving
 The processors share common system clock and common data structures.

15. What is the main advantage of a clustered system?


Storage area networks (SANs), which allows many systems to attach to a pool of storage.

16. Explain symmetric and asymmetric clustering.

3
 In symmetric clustering, two or more hosts are running applications and are monitoring
each other. This structure is obviously more efficient, as it uses all of the available
hardware. However, it does require that more than one application be available to run.
 In asymmetric clustering, one machine is in hot-standby mode while the other is
running the applications.

17. What is software paralellization?


Parallel processing software is a mid-tier application that manages program task execution
on a parallel computing architecture by distributing large application requests between
more than one CPU within an underlying architecture, which seamlessly reduces execution
time.

18. What is distributed lock manager (DLM)? Where is it used?


 Distributed lock manager (DLM) is a foundation that the operating systems use to
organize and serialize the access to resources.
 Distributed lock manager runs in every machine in a cluster, with an identical copy of a
cluster-wide lock database.

You might also like