wayground logo

Free Printable Worksheets

NEW

Font size

S
M
L
XL
Worksheets

The Fallingwater Analogy

Total questions: 62

Worksheet time: 31mins

Name
Class
Date
1.

Which idea best captures principled design when using the Fallingwater analogy in IT systems?

a)

Changing goals repeatedly during implementation

b)

Adhering to core design principles consistently

c)

Prioritizing aesthetics over system reliability

d)

Maximizing features regardless of constraints

2.

In architecture and computing, organic architecture suggests what design approach?

a)

Using rigid modular rules without exceptions

b)

Separating components to avoid interactions

c)

Focusing only on performance benchmarks

d)

Integrating system with its environment harmoniously

3.

A design tradeoff highlighted by the Fallingwater analogy would most likely involve which choice?

a)

Ignoring site conditions to simplify construction

b)

Avoiding any compromises to meet perfection

c)

Selecting every possible feature without limits

d)

Balancing aesthetic integration with structural constraints

4.

You must propose a system architecture for a workload with frequent memory row conflicts. Which principle-guided strategy aligns with the analogy?

a)

Delay decisions until hardware issues disappear

b)

Choose an ISA and microarchitecture that mitigate contention

c)

Add more unrelated features to impress stakeholders

d)

Rely solely on compiler optimizations for fixes

5.

When applying creative thinking to IT design inspired by Fallingwater, what is a practical step?

a)

Generate solutions that fit constraints and context

b)

Ignore limitations to brainstorm without boundaries

c)

Copy existing designs without adaptation

d)

Focus on one metric and disregard others

6.

Which statement best distinguishes design principles from design precedents in IT systems?

a)

Principles are opinions of designers; precedents are regulations

b)

Principles are universal rationales; precedents are past solutions

c)

Principles are vendor practices; precedents are scientific laws

d)

Principles are old examples; precedents are timeless theories

7.

A team faces cache interference across cores. Which approach reflects out-of-the-box thinking grounded in principles rather than copying precedents?

a)

Replicate another vendor’s cache hierarchy exactly

b)

Analyze memory contention at each layer, redesign controller policy

c)

Adopt the most cited scheduling algorithm unchanged

d)

Increase DRAM capacity following typical upgrades

8.

Breaking abstraction layers can help solve performance issues. Which action applies this concept to the DRAM interface shown?

a)

Rely solely on operating system defaults for memory

b)

Only optimize application code without hardware insight

c)

Ignore lower-level behavior and add more cores

d)

Profile access patterns, tune controller-to-bank mapping

9.

Which component pair forms the fundamental elements of a DRAM cell, as shown in the diagram?

a)

Two capacitors in parallel

b)

A capacitor and access transistor

c)

Two access transistors in series

d)

A flip-flop and buffer gate

10.

Why must each DRAM row be periodically refreshed?

a)

To rebalance load across ranks

b)

Capacitor charge leaks over time

c)

Wordlines drift due to thermal noise

d)

To rewrite corrupted parity bits

11.

A system has 1 Exabyte of DRAM with 8 KB rows. Approximately how many rows does it contain?

a)

About 1.34e11 rows

b)

About 1.25e16 rows

c)

About 1.34e14 rows

d)

About 1.25e8 rows

12.

Which downside directly affects performance predictability during DRAM refresh?

a)

Energy per refresh increases heat

b)

Ranks unavailable cause pause times

c)

Higher leakage reduces retention time

d)

Refresh limits capacity scaling

13.

Which statement best captures the multi-directional focus of a computer architect?

a)

Looking backward, forward, up, and down across computing

b)

Looking only forward to future device technologies

c)

Looking upward at applications and software stacks

d)

Looking downward to circuits and manufacturing details

14.

In the computing stack, which pair correctly represents looking up and looking down for a computer architect?

a)

Up: DRAM cells; Down: operating systems

b)

Up: circuits; Down: compilers

c)

Up: instruction set; Down: user requirements

d)

Up: applications; Down: device technology

15.

From the performance chart, what trend is shown as DRAM device capacity increases?

a)

Higher percentage of time spent refreshing

b)

Refresh time peaks at mid-range capacity only

c)

Lower percentage of time spent refreshing

d)

Unchanged refresh time percentage across capacities

16.

Which rationale explains why refresh overhead becomes a design tradeoff for future DRAM systems?

a)

Refresh reduces error rates as capacity increases

b)

Refresh overhead is dominated by CPU arithmetic operations

c)

Refresh consumes more performance and energy at larger capacities

d)

Refresh becomes unnecessary due to perfect retention

17.

What question on the problem-solving slide encourages architectural innovation about DRAM refresh?

a)

Must buses always be synchronous for reliability?

b)

Is virtual memory needed for embedded systems?

c)

Do we need to refresh all rows every 64ms?

d)

Should caches be write-through at all levels?

18.

Which strategy aligns with exposing lower-level information to upper layers to reduce refresh overhead?

a)

Retention-aware refresh using device-level knowledge

b)

Uniform periodic refresh regardless of cell behavior

c)

Increasing CPU frequency to mask memory stalls

d)

Ignoring device variability to simplify controllers

19.

From the energy chart, what approximate fraction of DRAM energy may be spent on refresh at 64 Gb capacity?

a)

About fifteen percent of total DRAM energy

b)

About eight percent of total DRAM energy

c)

About forty-seven percent of total DRAM energy

d)

About twenty-six percent of total DRAM energy

20.

Which duty best represents "looking backward" in an architect’s role?

a)

Mapping APIs to hardware features

b)

Forecasting emerging workload trends

c)

Selecting future fabrication processes

d)

Analyzing historical design trade-offs

21.

What does "looking forward" primarily require from a systems architect?

a)

Minimizing compiler optimization barriers

b)

Auditing current operational incidents

c)

Documenting past performance regressions

d)

Predicting technology and workload evolution

22.

In "looking up/down," what interface is most central to bridge effectively?

a)

Physical data center cooling methods

b)

Hardware/software boundary and contracts

c)

User interface accessibility guidelines

d)

Network transport layer protocols

23.

Which set best captures system design goals an architect balances?

a)

Aesthetics, branding, licensing, layout

b)

Monetization, advertising, virality, reach

c)

Performance, efficiency, reliability, cost

d)

Gamification, badges, leaderboards, scores

24.

When deciding how much low-level profile information to expose, what trade-offs arise?

a)

Marketing, hiring, training, travel

b)

Logging, alerting, paging, reporting

c)

Overhead, power, verification, cost

d)

Caching, paging, swapping, hashing

25.

To whom might retention-time profile information be exposed for best system impact?

a)

DFS, CDN, SEO, CRM subsystems

b)

Memory controller, OS, programmer, compiler

c)

Printers, scanners, cameras, displays

d)

End users, sales team, HR, finance

26.

Which reasoning best justifies parallelizing DRAM refreshes with accesses?

a)

Serialize accesses to ensure strict consistency

b)

Disable refresh to save dynamic power always

c)

Overlap refresh latency to improve throughput

d)

Increase refresh duration to reduce wear-out

27.

Which statement best describes the role of abstraction layers in computing systems?

a)

They hide low-level details to manage complexity

b)

They eliminate hardware to simplify programming

c)

They guarantee performance regardless of workload

d)

They convert analog signals into digital signals

28.

In the hierarchy from problems to electrons, where does the Instruction Set Architecture (ISA) primarily operate?

a)

Inside the compiler optimization stage

b)

Between software and microarchitecture boundary

c)

Above high-level application requirements

d)

Below transistor fabrication processes

29.

What is microarchitecture responsible for relative to the ISA?

a)

Regulating power by operating system policies

b)

Implementing ISA operations using hardware structures

c)

Defining the programmer-visible instruction formats

d)

Translating high-level languages into machine code

30.

Which action exemplifies breaking abstraction layers to design better systems?

a)

Relying solely on compiler defaults for performance

b)

Using hardware retention profiling to tune refresh rates

c)

Writing code only against standard library interfaces

d)

Ignoring cache behavior to keep code portable

31.

RAIDR’s key idea for DRAM refresh can be summarized as what policy?

a)

Refresh weak rows more often, strong rows less often

b)

Refresh all rows uniformly at the shortest interval

c)

Duplicate rows to avoid retention-time profiling

d)

Disable refresh to reduce dynamic power usage

32.

Which step in RAIDR stores rows by retention time using compact metadata?

a)

Binning with Bloom Filters in the controller

b)

Profiling with operating system timers

c)

Encoding with error-correcting codes

d)

Refreshing with fixed auto-refresh commands

33.

What phenomenon is illustrated when repeatedly activating a DRAM aggressor row causes bit flips in neighbors?

a)

Instruction-level hazards in pipelines

b)

Disturbance errors in adjacent victim rows

c)

Write-back stalls in cache hierarchy

d)

Voltage droop in power delivery network

34.

Why can cooperation across components and layers yield more effective systems?

a)

Cross-layer information enables targeted optimizations

b)

More layers guarantee fewer bugs by design

c)

Adding interfaces always reduces energy consumption

d)

Ignoring lower levels avoids complexity in solutions

35.

Which concept best distinguishes productivity from efficiency in IT system design?

a)

Maximizing output per unit resource consumed

b)

Doing more tasks regardless of resource cost

c)

Using the newest hardware without optimization

d)

Adding more software layers for modularity

36.

Why is crossing abstraction layers important when dealing with DRAM disturbance errors shown across CPU architectures?

a)

It enables coordinated fixes across hardware and software

b)

It simplifies documentation for developers

c)

It removes the need for error monitoring

d)

It guarantees zero errors on all platforms

37.

Which action is an example of tolerating memory disturbance errors at the hardware/software interface?

a)

Throttle CPUs to reduce access rates only

b)

Ignore errors and rely on user reports

c)

Use stronger error correcting codes proactively

d)

Replace DRAM with a different technology

38.

Given the access rates and error counts, which inference supports a performance–reliability trade-off across architectures?

a)

Access rate is irrelevant to reliability planning

b)

Higher access rates always eliminate errors

c)

Errors are independent of access behavior

d)

Lower error counts can occur at moderate access rates

39.

Which approach best embodies embracing heterogeneity across abstraction layers to manage disturbance errors?

a)

Disable caching to force uniform memory reliability

b)

Standardize all modules to identical reliability levels

c)

Move all computation to external storage devices

d)

Map error-tolerant data to less reliable memory sections

40.

In a multi-core processor, which shared resource is most likely to cause system-wide slowdowns when one thread becomes a memory performance hog?

a)

Dedicated integer ALUs

b)

Per-core branch predictors

c)

Shared last-level cache bandwidth

d)

Private L1 instruction caches

41.

Which scenario best illustrates resource contention in a multi-core system leading to unexpected performance drops?

a)

One core streaming large arrays saturating memory bandwidth

b)

Each core running short control-heavy tasks

c)

Two cores executing independent integer arithmetic

d)

Cores idling while waiting for IO interrupts

42.

What is the primary effect when interconnect contention rises due to a memory-intensive workload on one core?

a)

Reduced branch prediction accuracy across cores

b)

Improved locality in private L1 caches

c)

Increased latency for cache-to-memory transactions

d)

Higher throughput of floating-point pipelines

43.

A four-core system has a shared last-level cache and a ring interconnect. One core runs a bandwidth-heavy copy loop. Which mitigation is most likely to reduce cross-core slowdowns without changing the algorithm?

a)

Pin the copy loop to a core closer to memory controller

b)

Disable private L1 caches on all cores

c)

Increase integer ALU issue width per core

d)

Lower the branch predictor table size

44.

When a single thread becomes a memory performance hog, which symptom would you expect in the other cores’ memory access profiles?

a)

More stalled cycles waiting on interconnect

b)

Faster commit rates in integer pipelines

c)

Shorter average memory access times

d)

Lower miss rates at all cache levels

45.

Which design choice can help isolate heavy memory traffic and reduce shared resource contention in multi-core systems?

a)

Unified execution pipeline shared across threads

b)

Global branch predictor shared by all cores

c)

Per-core private last-level caches with no sharing

d)

Partitioned or QoS-controlled shared cache and memory bandwidth

46.

In a DRAM bank diagram, which component caches the most recently activated row to serve subsequent accesses quickly?

a)

Row buffer inside bank

b)

Command queue in controller

c)

Column decoder buffer

d)

Global address latch

47.

A memory access that targets the currently open row in a DRAM bank is called what?

a)

Row-conflict access

b)

Row-skip access

c)

Row-hit access

d)

Row-miss access

48.

Which sequence best describes serving a row-conflict in a DRAM bank?

a)

Refresh then issue WRITE

b)

Issue READ immediately

c)

Precharge then activate new row

d)

Activate then precharge row

49.

In FR-FCFS scheduling, which request is prioritized first among ready requests?

a)

Oldest request regardless of type

b)

Row-hit ready request

c)

Highest address request

d)

Writeback cleanup request

50.

What is the main unfairness risk of FR-FCFS scheduling under mixed workloads?

a)

Starves refresh operations entirely

b)

Ignores bank-level parallelism

c)

Prefers writes over reads always

d)

Keeps row-hit stream dominating others

51.

Which action closes the currently open row in a DRAM bank to prepare for activating another row?

a)

Precharge command

b)

Burst terminate

c)

Activate command

d)

Refresh command

52.

When multiple banks exist, which policy increases throughput by exploiting parallelism while minimizing conflicts?

a)

Round-robin per core

b)

Bank-aware FR-FCFS

c)

Close-page on every access

d)

Always-open row policy

53.

Which timing constraint limits how quickly a controller can issue an ACT to a different row in the same bank after a PRE?

a)

tRCD constraint

b)

tRAS constraint

c)

tWR constraint

d)

tRP constraint

54.

A controller that batches writes to drain the write queue primarily aims to reduce which cost?

a)

Row activation energy

b)

Read-to-write turnaround

c)

Address bus contention

d)

Bank refresh overhead

55.

Which memory access pattern typically maximizes row buffer locality in DRAM and thus improves throughput under common scheduling policies?

a)

Uniformly distributed accesses with large strides

b)

Interleaved accesses alternating rows per request

c)

Streamed accesses within the same open row

d)

Random accesses across many banks and rows

e)

Bursty accesses with irregular stride

56.

In a memory performance attack, which condition most directly leads to denial of service for a victim application?

a)

Victim has low cache hit rate only

b)

Attacker monopolizes row buffer hits

c)

Controller disables refresh operations

d)

Victim uses multiple memory channels

e)

Attacker reduces precharge timing

57.

Why can random access patterns exploit unfairness in certain DRAM controllers?

a)

They make refresh invisible to scheduling

b)

They increase page hits for all threads

c)

They reduce command bus utilization

d)

They force frequent precharges and row conflicts

e)

They avoid bank-level parallelism entirely

58.

Which factor best defines memory intensity when analyzing susceptibility to performance attacks?

a)

Depth of the reorder buffer

b)

Proportion of time stalled on memory

c)

Number of cores in the processor

d)

Frequency of branch mispredictions

e)

Size of the instruction cache only

59.

A DRAM controller favoring row-buffer hits with FR-FCFS scheduling is most likely to:

a)

Prioritize random requests with long latency

b)

Throttle all threads equally under load

c)

Close the row after every single access

d)

Prioritize requests that hit the open row

e)

Prioritize oldest requests regardless of row state

60.

An attacker issues a stream of accesses to one hot row while a victim issues random accesses. What is the expected outcome?

a)

Controller grants attacker priority causing victim slowdowns

b)

Attacker suffers more precharges than the victim

c)

Victim experiences more row hits and higher throughput

d)

Both threads get equal service due to fairness

e)

Row buffer thrashing eliminates any advantage

61.

Which mitigation most directly reduces unfairness caused by row buffer locality exploitation?

a)

Increase DRAM refresh rate significantly

b)

Adopt request prioritization caps per thread

c)

Disable bank interleaving entirely

d)

Use larger last-level caches only

e)

Always close rows immediately after activation

62.

When evaluating a suspected memory denial-of-service scenario, which evidence best supports the diagnosis?

a)

High CPU utilization with few memory stalls

b)

Lower queue occupancy in the controller

c)

Victim’s row-hit rate plummets under attacker load

d)

Reduced bank-level parallelism for the attacker

e)

Stable instruction throughput across all threads