A dictionary consists of english words stored in memory. Maximum number of words in the dictionary is 1M.
Write a command-line program that allows concurrent (by multiple processes):
- search for existence of a word in the dictionary
- insert a word in the dictionary, if it does not exist
- delete a word from the dictionary
The command should be invoked as following:
dict {insert|search|delete} <word>
Languages to chose from: C or C++ Assume APIs for data structures like queue, hash table, linked list etc. are available. Document the APIs (no implementation required) as comments.
Based on lock queue information from individual nodes, detect deadlocks and chose which query to kill in order to resolve it most effectively.
Sample input (example is in json but you can adjust the format to suite the language of choice):
{
"node1": { "obj1": [1, 3], "obj2": [4, 10] }
"node2": { "obj1": [5, 4] }
"node3": { "obj2": [3, 5] }
}
Meaning that on node1 the query 1 is holding lock on obj1 and query 3 is waiting for it, query 4 is holding lock on obj2 and query 10 is waiting for it, etc.
The queryIds represent a distrbuted queries so one query will get same queryId on all nodes. Object names are unique across cluster.
You can assume max 1000 nodes and max 1000 queries on each node.
Languages to chose from: C, C++, Python, Perl, nodejs, SQL Assume APIs for data structures like queue, hash table, linked list etc. are available. Document the APIs (no implementation required) as comments.