ATTA Sample Questions & Answers

The response time is well within the required limit (180ms < 200ms), but the throughput is significantly below the target (350 TPS < 500 TPS). The critical indicator is the CPU utilization at 95%, which means the processor is maxed out and cannot handle more work. This CPU saturation is preventing the system from achieving the target throughput. Memory utilization is low, and there is no data to suggest a network or database issue.

Decision coverage is a control-flow testing technique that ensures all branches in the code are exercised. Data flow analysis, which detects UR anomalies, focuses on the lifecycle of variables. It is possible to have 100% decision coverage while still having a data path where a variable is used before initialization. For example, the use of the uninitialized variable could occur within a code block that is fully executed as part of a single branch.

To show that 'B' independently affects the outcome, 'B' must be the only condition that changes, the other conditions must be held constant, and the overall outcome of the decision must change. In this pair, A is held as True, C is held as False. When B changes from True to False, the outcome changes from True (T && (T || F) -> T) to False (T && (F || F) -> F). This is the correct independence pair for condition B.

Data caching inherently introduces a potential for stale data, which is a consistency problem. While caching significantly improves performance by serving data from a local, faster copy, it creates a challenge in ensuring that the cached data is invalidated or updated when the source data changes. In a system where data consistency is the top priority, the caching strategy (e.g., write-through, write-back, time-to-live) must be rigorously reviewed.

This attack works by injecting SQL commands into the input field. The ' OR 1=1 clause modifies the WHERE clause of the SQL query to always be true, and the -- comments out the rest of the original query (like the password check). This allows the attacker to bypass the intended logic of the SQL statement, which is the definition of an SQL Injection attack.

Fault injection is the process of deliberately introducing errors or failures into a system to test its response. Terminating a server instance is a classic example of injecting a fault into the system's environment to verify its fault tolerance and recoverability mechanisms. This practice is often referred to as Chaos Engineering.

Cyclomatic Complexity can be calculated using the formula M = E - N + 2P, where E is the number of edges, N is the number of nodes, and P is the number of connected components (P=1 here). The graph has 11 nodes (A-K) and 14 edges. So, M = 14 - 11 + 2 = 5. Alternatively, using the simpler formula M = D + 1, where D is the number of decision nodes (nodes with more than one outbound edge). The decision nodes are B, C, F, and G. There are 4 decision nodes. So, M = 4 + 1 = 5.

Coexistence testing, a sub-characteristic of compatibility, verifies that an application can run in the same environment (e.g., on the same hardware and OS) with other applications without causing resource conflicts or other negative interactions. Integrating a new library version falls directly under this, as it must coexist peacefully with all other software components in the environment.

The number of decision points determines the cyclomatic complexity and the minimum number of paths for path coverage, which is related to the number of tests for decision or branch coverage. However, statement coverage simply requires that every executable line of code is run at least once. A single, carefully constructed test case could potentially execute all statements in the module if there's a linear path through all branches. Conversely, it might take many test cases. The number of decision points does not directly determine the number of tests for statement coverage.

While several security characteristics are relevant, Accountability is the most specific. It is defined as the degree to which the actions of an entity can be traced uniquely to the entity. Testing for this requirement involves ensuring the system correctly checks that the user associated with the session token has the proper authorization (rights and privileges) to perform the 'execute trade' action. This links the action directly to an accountable, authorized user.