ISRO Scientist/Engineer Computer Science 2025 paper: Part A, questions 61 to 80, solved
Questions 61 to 80 of the ISRO Scientist/Engineer 'SC' Computer Science written test of 26 Oct 2025 (Set A): 20 questions, each solved by Skill-Lync and checked against ISRO's final answer key.
Question paper and answer keys published by ISRO's Centralised Recruitment Board; solutions by Skill-Lync · Updated 18 Nov 2026 · 20 of 80 questions on this page
Number of elements in A^n for a set A with m elements
- (a)
- (b)
- (c)
- (d)
Answer: (d) Official final key: (d)
( factors) is the set of ordered -tuples with each entry from . Each of the positions can be filled in ways independently, so
For example , gives pairs, while would give . Option (c) swaps the base and the exponent.
The same counting gives the size of a state space: bits give states, and a record of fields with values each has combinations to test.
Four-letter words from GAGANYAAN with no repeated letter
- (a)3024
- (b)192
- (c)64
- (d)24
Answer: (d) 24 Official final key: (d)
GAGANYAAN has 9 letters but only four distinct ones: G (twice), A (four times), N (twice) and Y (once). With no letter repeated in the word, each of G, A, N, Y can be used at most once, so a four-letter word must use all four.
The number of arrangements of four different letters is
The figure 3024 is , which wrongly treats the nine letters as all distinct.
24
Counting arrangements with repeated symbols is the basis for sizing the search space of identifiers, codes and test-case permutations.
Situation where the outcome depends on the order of concurrent data access
- (a)Stack overflow
- (b)Aging
- (c)Race Condition
- (d)Buffer overflow
Answer: (c) Race Condition Official final key: (c)
This is the textbook definition of a race condition. If two processes each run counter = counter + 1 as read, add, write, the order of those steps decides the result: run one after the other and the counter goes up by 2, interleave the two reads first and it goes up by only 1.
The others are different things: stack overflow is running out of stack space, aging is a scheduling technique that raises the priority of waiting processes to avoid starvation, and buffer overflow is writing past the end of a buffer.
Race Condition
Race conditions in shared telemetry buffers or between interrupt handlers and tasks cause rare, hard-to-reproduce faults in flight and ground software; mutexes, atomics and message queues are the usual fixes.
What a parent process does when it creates a new child process
- (a)The parent continues to execute concurrently with its children
- (b)The parent waits until some or all of its children have terminated
- (c)The parent waits for the child process by issuing a signal ( ) system call
- (d)Both (a) and (b)
Answer: (a) The parent continues to execute concurrently with its children Official final key: (a) · provisional key was (d)
When a process creates a child (for example with fork()), the operating system lets both run. By default the parent continues to execute concurrently with its children, so (a) is true.
The parent waits only if it chooses to, by calling wait(). Creating a process does not by itself make the parent wait, so (b) is not what happens on creation, and (d) fails for the same reason. Option (c) names the wrong call: waiting is done with wait(), not signal().
Textbooks list both behaviours as choices a parent can make; ISRO’s final key treats concurrent execution as the default, and the parent waits only if it calls wait().
The parent continues to execute concurrently with its children
ISRO's provisional key (Oct 2025) gave (d); the December 2025 key changed it to (a), and the final key of 15 May 2026 kept (a).
Supervisor and launcher processes on ground systems spawn workers and either monitor them concurrently or block on them with wait(), and choosing wrongly leaves zombie processes or stalled pipelines.
Which process state transition is not possible (Run, Ready, Waiting, Terminated)
Which of the following transitions is not possible between the process states shown below?
- (a)Run Ready
- (b)Run Terminated
- (c)Run Waiting
- (d)Waiting Run
Answer: (d) Waiting Run Official final key: (d)
In the standard process state diagram, a process moves between New, Ready, Running, Waiting (blocked) and Terminated. Only the scheduler's dispatcher moves a process into Running, and it always picks from the Ready queue.
- Run Ready: possible, the time slice expires or a higher-priority process pre-empts it.
- Run Terminated: possible, the process exits.
- Run Waiting: possible, the process asks for I/O or waits for an event.
- Waiting Run: not possible. When the I/O or event completes, the process goes Waiting Ready and must wait to be dispatched again.
Waiting Run is the impossible transition.
Every RTOS and Linux scheduler enforces exactly this state machine; a task that wakes from an interrupt is made ready, never directly run, which is why wake-up latency is measured as the time to reach the CPU after becoming ready.
Wait-for graph corresponding to a resource allocation graph with P1 to P5 and R1 to R5
Analyze the below resource allocation graph,
Which of the following is the corresponding wait-for graph?
- (a)
- (b)
- (c)
- (d)
Answer: (a) Graph (a) Official final key: (a)
A wait-for graph is made from a resource allocation graph (single instance of each resource) by deleting the resource nodes: draw an edge when requests a resource that holds, that is, when the RAG has .
Reading the RAG: gives . gives . gives . gives . gives . gives .
So the wait-for graph has six edges: , , , , , . Graph (b) lacks , (c) lacks , and (d) lacks . Only graph (a) has all six.
Graph (a)
Databases and distributed lock managers build exactly this graph at run time and search it for cycles to detect deadlock, then abort a victim transaction to break the cycle.
Average turnaround and waiting time under FCFS for P1, P2, P3
For the following list of processes, the average turnaround time and average waiting time using FCFS algorithm is :
| Process | Arrival Time (ms) | Execution Time (ms) |
|---|---|---|
| P1 | 6 | 24 |
| P2 | 0 | 3 |
| P3 | 3 | 3 |
- (a)Average Turnaround time 30 ms and Average Waiting Time 4.5 ms
- (b)Average Turnaround time 13 ms and Average Waiting Time 3 ms
- (c)Average Turnaround time 10 ms and Average Waiting Time 0 ms
- (d)Average Turnaround time 24 ms and Average Waiting Time 0 ms
Answer: (c) Average Turnaround time 10 ms and Average Waiting Time 0 ms Official final key: (c)
FCFS serves processes in order of arrival: (arrives 0), (arrives 3), (arrives 6).
runs 0 to 3, runs 3 to 6 and runs 6 to 30. Each process starts exactly when it arrives, so nobody waits.
Turnaround completion arrival: , , . Average .
Waiting turnaround burst for all three, so the average is .
and
Turnaround and waiting time are the standard first checks when comparing batch-job schedulers, for example in a ground-station processing queue where jobs are served in arrival order.
Identify wait, run, turnaround and response time from a scheduling timeline
In the context of Operating system process scheduling, if W represents Wait Time, R represents Run Time, T represents Turnaround time, S represents Response time, then which of the following is correct option with respect to given diagram?
- (a)Time1:W, Time2:R, Time3:T, Time4:S
- (b)Time1:S, Time2:W, Time3:R, Time4:T
- (c)Time1:W, Time2:R, Time3:S, Time4:T
- (d)Time1:R, Time2:S, Time3:W, Time4:T
Answer: (c) Time1:W, Time2:R, Time3:S, Time4:T Official final key: (c)
Read each bar against the marks on the time axis (Arrival, Execution, First Output, Finish).
- Time1 runs from Arrival to Execution: the job sits in the queue before it starts, so it is the wait time W.
- Time2 runs from Execution to Finish: the time the job is actually running, so it is the run time R.
- Time3 runs from Arrival to First Output: the time before the user first sees a response, so it is the response time S.
- Time4 runs from Arrival to Finish: the total time in the system, so it is the turnaround time T.
That is Time1:W, Time2:R, Time3:S, Time4:T. Option (a) swaps S and T, (b) and (d) mislabel the first bar.
(c) Time1:W, Time2:R, Time3:S, Time4:T
Latency dashboards and scheduler tracing tools (for example Linux scheduling traces) report exactly these four quantities per task, and mixing up response and turnaround time is a common cause of wrong tuning.
Banker's algorithm: is the snapshot of P0 to P2 over resources A, B, C safe
Consider a system with 3 processes P0 to P2 and 3 resource types A, B, and C. Resource type A has 10 instances, resource type B has 5 instances and resource type C has 7 instances. The following snapshot of the system has been taken at a given time.
| Allocation A | B | C | Max A | B | C | Available A | B | C | |
|---|---|---|---|---|---|---|---|---|---|
| P0 | 0 | 1 | 0 | 7 | 5 | 3 | 5 | 4 | 5 |
| P1 | 2 | 0 | 0 | 3 | 2 | 2 | |||
| P2 | 3 | 0 | 2 | 9 | 0 | 2 |
Which of the following is true?
- (a)The system is not in safe state
- (b)The system is blocked
- (c)The system is in safe state
- (d)The request cannot be granted to all the processes
Answer: (c) The system is in safe state Official final key: (c)
Safety check: . This gives , , . Start with .
- : need , so it can finish; Work becomes .
- : need , so it can finish; Work becomes .
- : need , so it can finish; Work becomes .
The sequence lets every process finish, so the state is safe. Nothing is blocked, and the snapshot contains no pending request, so (b) and (d) do not follow.
(c) The system is in safe state
Resource-ordering and admission checks inspired by the Banker's algorithm appear in schedulers and memory or connection-pool managers that must refuse a request which could lead to deadlock.
Page faults under Optimal replacement with 3 frames, string 6,0,5,2,0,3,0,4,2,3,0,3,2,5,2
Consider the following reference strings and assume 3 frames are available, then the number of page faults that occur using Optimal page replacement algorithm is
6, 0, 5, 2, 0, 3, 0, 4, 2, 3, 0, 3, 2, 5, 2
- (a)15 Page Faults
- (b)9 Page Faults
- (c)12 Page Faults
- (d)8 Page Faults
Answer: (d) 8 Page Faults Official final key: (d)
Optimal replacement evicts the page whose next use is farthest in the future (or never).
- 6, 0, 5 fill the three frames: 3 faults.
- Page 2 faults: 6 is never used again, so evict 6. Frames . (4 faults)
- Page 0 hits. Page 3 faults: next uses are 2 at position 9, 0 at position 7, 5 at position 14, so evict 5. Frames . (5 faults)
- Page 0 hits. Page 4 faults: next uses are 2 (9), 3 (10), 0 (11), so evict 0. Frames . (6 faults)
- Pages 2 and 3 hit. Page 0 faults: 4 is never used again, so evict 4. Frames . (7 faults)
- Pages 3 and 2 hit. Page 5 faults: 0 and 3 are never used again, evict either. (8 faults)
- The last page, 2, hits.
Total page faults.
page faults
Real operating systems cannot use Optimal because it needs the future, but it is the benchmark: cache and page-replacement policies such as LRU or CLOCK are evaluated by how close their fault counts come to it on recorded traces.
Cyclomatic complexity is associated with which testing approach
Cyclomatic complexity is associated with
- (a)White Box Testing
- (b)Black Box Testing
- (c)Grey Box Testing
- (d)None of the above
Answer: (a) White Box Testing Official final key: (a)
Cyclomatic complexity is computed from the control-flow graph of the code: for a connected graph with edges and nodes, which equals the number of decision points plus one. It tells you the number of independent paths through the code and therefore the minimum number of test cases needed to cover every path (basis path testing).
Because it needs the internal structure of the code, it belongs to white box (structural) testing. Black box testing works only from the specification.
(a) White Box Testing
Coding standards for safety-critical embedded and flight software commonly cap cyclomatic complexity per function, and static-analysis tools report it on every build.
Which testing is related to Alpha Testing and Beta Testing
Which of the following is related to Alpha Testing and Beta Testing?
- (a)Unit Testing
- (b)System Testing
- (c)Acceptance Testing
- (d)Integration Testing
Answer: (b) System Testing or (c) Acceptance Testing Official final key: (b or c), either accepted · provisional key was (c)
Alpha testing is done by users or testers inside the developing organisation, and beta testing is done by a limited set of real users in their own environment, both on the finished product before general release. In the standard textbook classification these are the two forms of acceptance (user acceptance) testing. Unit testing checks single modules and integration testing checks how modules fit together, so (a) and (d) are wrong.
ISRO's final key accepts both (b) and (c). The key gives no reason; the textbook reading is (c). Alpha and beta runs also exercise the complete, integrated product, which is what system testing means, so a case for (b) can be made, but (c) is the answer to give in any exam that wants one letter.
(c) Acceptance Testing (ISRO also accepts (b) System Testing)
ISRO's provisional key (Oct 2025) gave (c); the December 2025 key changed it to "b or c", and the final key of 15 May 2026 kept "b or c".
Release pipelines use staged rollouts that follow this idea: an internal alpha build for the team, then a beta or canary build for a small set of real users, before the build reaches everyone.
Capability Maturity Model process maturity levels 2, 3 and 4
As per the Capability Maturity Model (CMM), the process maturity levels are defined as
Level 1 : Initial
Level 2 : (A)
Level 3 : (B)
Level 4 : (C)
Level 5 : Optimizing
Identify (A), (B) and (C).
- (a)A : Defined, B : Repeatable, C : Managed
- (b)A : Repeatable, B : Managed, C : Defined
- (c)A : Repeatable, B : Defined, C : Managed
- (d)A : Defined, B : Managed, C : Repeatable
Answer: (c) A : Repeatable, B : Defined, C : Managed Official final key: (c)
The five CMM levels in order are: 1 Initial (ad hoc), 2 Repeatable (basic project management, earlier success can be repeated), 3 Defined (a documented, standard process across the organisation), 4 Managed (process and products are measured quantitatively), 5 Optimizing (continuous improvement).
So (A) is Repeatable, (B) is Defined and (C) is Managed.
(c) A : Repeatable, B : Defined, C : Managed
Process-maturity ratings such as CMMI levels are a common requirement in contracts for mission software and in vendor audits for space and defence work.
Correct order of stages in the software risk management process
As per software engineering, which of the following order is correct for stages of Risk Management process of a software project?
- (a)Risk identification Risk planning Risk analysis Risk monitoring
- (b)Risk planning Risk identification Risk analysis Risk monitoring
- (c)Risk analysis Risk identification Risk planning Risk monitoring
- (d)Risk identification Risk analysis Risk planning Risk monitoring
Answer: (d) Risk identification Risk analysis Risk planning Risk monitoring Official final key: (d)
The standard risk management process is: first identify the possible risks, then analyse each one (how likely it is and how bad it would be), then plan how to avoid or reduce it, and finally keep monitoring the risks through the project.
You cannot analyse or plan for a risk you have not identified, and you plan only after you know which risks matter. That gives identification, analysis, planning, monitoring.
(d) Risk identification Risk analysis Risk planning Risk monitoring
Project risk registers in space and embedded programmes follow this cycle: log a risk, score it by likelihood and impact, assign a mitigation, and review it at each milestone.
Fan-in and fan-out of a function X: coupling and complexity statements
Considering, X is a function defined in a software project, following statements are made
- Fan-in is a measure of the number of functions that call the function, X
- Fan-out is the number of functions which are called by the function, X
- High value of Fan-in means that X is tightly coupled to the rest of design of software
- High value of Fan-out means that overall complexity of X may be high
Which of the following options is correct for the above-mentioned statements?
- (a)Only (i) and (ii) are true
- (b)Only (i) and (iii) are true
- (c)Only (ii) and (iv) are true
- (d)(i), (ii), (iii) and (iv) are true
Answer: (d) (i), (ii), (iii) and (iv) are true Excluded from evaluation by ISRO
Statements (i) and (ii) are the definitions: fan-in counts the functions that call X, fan-out counts the functions X calls.
Statements (iii) and (iv) are the usual textbook readings (for example in Sommerville's design-quality discussion): a high fan-in means many parts of the program depend on X, so X is tightly coupled to the rest of the design and a change to it ripples widely; a high fan-out means X has to coordinate many other functions, so the control logic inside X, and hence its complexity, may be high. All four statements are therefore true, which is option (d).
ISRO excluded this question from evaluation "for administrative reasons". Its provisional key (Oct 2025) gave (d), the December 2025 key accepted (a) or (d), and the final key of 15 May 2026 excludes it. ISRO gave no reason; the wording of (iii) and (iv) is where the debate lies, since a high fan-in is also sometimes read as good reuse. By the standard definitions above, (d) is the technically correct choice.
(d) (i), (ii), (iii) and (iv) are true (technical answer; the question carries no marks)
Static-analysis and architecture tools report fan-in and fan-out per function or module; a module with very high fan-in is a risky place to change, and one with very high fan-out is a candidate for splitting.
V Model of software testing: identify the test levels A, B, C, D
In the V Model of Software Testing shown below, please identify (A), (B), (C), (D)
- (a)(A): Acceptance Testing, (B): System Testing, (C): Integration Testing, (D): Unit Testing
- (b)(A): Acceptance Testing, (B): Integration Testing, (C): System Testing, (D): Unit Testing
- (c)(A): Unit Testing, (B): System Testing, (C): Integration Testing, (D): Acceptance Testing
- (d)(A): Unit Testing, (B): Integration Testing, (C): System Testing, (D): Acceptance Testing
Answer: (a) (A): Acceptance Testing, (B): System Testing, (C): Integration Testing, (D): Unit Testing Official final key: (a)
In the V model each development phase on the left has a test phase at the same height on the right that verifies it.
- Requirement Analysis is checked by acceptance testing, so (A) is Acceptance Testing.
- System Design is checked by system testing, so (B) is System Testing.
- Architecture Design is checked by integration testing, so (C) is Integration Testing.
- Module Design is checked by unit testing, so (D) is Unit Testing.
The testing climbs back up from the smallest unit to the whole system, ending with the customer's acceptance.
(a) (A): Acceptance, (B): System, (C): Integration, (D): Unit Testing
Mission and flight software projects use V-model style plans to trace every requirement to a test level, so that unit, integration, system and acceptance tests each close a specific design document.
Function point value from inputs 30, outputs 60, enquiries 20, files 5, interfaces 4
For a software project, following information about the domain characteristics are given :
| Sl No | Description | Count | Average Weight factor |
|---|---|---|---|
| 1 | Number of user inputs | 30 | 4 |
| 2 | Number of user outputs | 60 | 5 |
| 3 | Number of user enquiries | 20 | 4 |
| 4 | Number of files | 5 | 10 |
| 5 | Number of external interfaces | 4 | 7 |
Assume that all complexity adjustment values are average having scale = 3. The function point value for the project will be
- (a)681.46
- (b)186.46
- (c)816.46
- (d)618.46
Answer: (d) 618.46 Official final key: (d)
Function points: , where is the weighted count and over the 14 complexity factors.
.
Every factor is 3, so and .
. Option (a) and (c) are digit shuffles of this value.
Function point counts are used to size a software contract before any code exists, and to compare effort and cost across projects regardless of the programming language.
Regular expression equivalent to (a*b)* + (b*a)* over {a, b}
Consider the regular expression over the alphabet :
Which of the following regular expressions below describe the same language as ?
- (a)
- (b)
- (c)
- (d)
Answer: (c) Official final key: (c)
generates the empty string and every string that ends in (cut the string after each ; each piece is some 's followed by a ). generates the empty string and every string that ends in (the same argument with the letters swapped).
Every non-empty string over ends in or in , so the union is every string, including the empty one: .
The others miss strings: (a) misses (it is one run of 's and 's only, in one order), (b) misses , and (d) misses the empty string and .
A brute-force check of every string up to length 10 agrees.
(c)
Regex equivalence reasoning is used when simplifying lexer rules and input validators; a pattern that quietly matches every string is a classic bug in log filters and URL routing.
Number of states in the minimal DFA for the given three-state automaton
Consider the deterministic finite-state automaton (DFA) below over the alphabet . The state with a small incoming arrow is the initial state, while a double circle state denotes a final state
The number of states in the MINIMAL DFA for the language accepted by this DFA is:
- (a)1
- (b)2
- (c)3
- (d)4
Answer: (b) 2 Official final key: (b)
Call the states (start, not final), (final, reached from on ) and (final, reached from on and from on ).
Start the partition with final versus non-final: and . From the symbols and lead to and , both final. From both and lead to , final. So and always land in the same block, and they cannot be told apart.
The automaton accepts every non-empty string: is left on the first symbol and the machine never returns to or to a non-final state. That language needs only one state to remember "nothing read yet" and one for "something read".
states
DFA minimisation is what lexer generators and regex engines do before emitting tables, so protocol parsers and log scanners run with the smallest possible state table.
Language generated by S -> aSb | T, T -> bT | b
Consider the context-free grammar below. Here is the starting non-terminal symbol, while and are terminal symbols.
The language generated by G is:
- (a)
- (b)
- (c)
- (d)
Answer: (a) Official final key: (a)
generates for : at least one .
Each use of adds one on the left and one on the right. After uses () and then , the string is with .
With this is with , i.e. . Option (b) wrongly demands (the string is generated with ); (c) has more 's than 's; (d) would need the rule to produce nothing.
Listing all derivable strings up to length 12 and comparing with each option confirms (a).
(a)
This is the reasoning behind checking grammars for protocol or file formats: a rule that forces extra closing symbols, such as more closing than opening brackets, defines a language no simple counter-free pattern can match.