Ab Initio Expert Quiz
1) We have two file A & B. A has data volume of 20 GB and B has only 10 MB and data is not sorted. If these two files to be joined, what you will recommend from below options :
- Select the Parameter sorted input to “input must be sorted” and make the File A as Deriving Port
- Select the Parameter sorted input to “input need not be sorted” and make the File A as Deriving Port.
- Select the Parameter sorted input to “input need not be sorted” and make the File B as Deriving Port.
- Select the Parameter sorted input to “input must be sorted” and make the File B as Deriving Port.
2) If you want to view the data in M-dump format through Abinitio component select
- view data formatted
- view data
- view data unformatted
- view data dump
3) To delete all watcher datasets in the default directory (for all graphs), In the GDE choose
- Debug > Delete Watcher Datasets
- Settings > Delete Watcher Datasets
- Debug > Remove All Watcher
- Settings > Delete All Watcher
4) You can verify the status of the graph in the end script using the variable
- $mpresult
- $mpjret
- $mpcheck
- $mpsuccess
5) Abinitio function cobol-to-dml converts REDIFINE of copybook into
- ReOrganize
- Organize
- Union
- Change
6) Which of the below functions is used for partitioned lookup files
- lookup_match
- lookup_partition
- lookup_local
- partition_lookup
7) m_eval ‘(decimal(6,2)) 1234’ gives
- 00
- 123400
- 00
- 1234
8) To read in a set of files based on a pattern in the filenames use below component?
- READ MULTIPLE INPUT
- READ MULTIPLE FILES
- READ MULTIPLE INPUT FILES
- MULTIPLE INPUT FILES
9) What does the error message “broken pipe” mean?
- data got corrupted
- pipe is present in the data
- downstream component has gone away unexpectedly
- Graph has aborted unexpectedly
10) Can I reverse-engineer a graph (that is, create an .mp file from a deployed .ksh script)?
- Yes
- No
- Maybe
- None of the above
11) Which function do we use to terminate a graph based on a condition?
- force_abort
- force_error
- force_terminate
- force_stop
12) What is the syntax used to locate files with URLs
- [ file | mfile ] :/directory1/directory2…/filename
- [ file | mfile ] ://hostname/directory1/directory2…/filename
- [ file | mfile ] ://servername/hostname/directory1/directory2…/filename
- [ file | mfile ] ://servername/directory1/directory2…/filename
13) The Graphs start script which performs setup actions is saved as
- .project-start.ksh
- .start-setup.ksh
- .start-graph.ksh
- .temp-start.ksh
14) When to opt for SED TYPE2?
- No history is required.
- History is required
- Only last modified information is required
- None of the above
15) If I am working in Abinitio and Teradata and I have a 1 TB Data that need to joined with existing data (4 TB) to get Delta records. What will you recommend :
- Get the data into one Temporary Table and do a join with existing data.
- Unload the existing data to Abinitio environment and use the Join component to get the delta.
- Create one Master File for existing data and do the Join with existing data.
- None of the above.
16) When to use Phase Break in Graph ?
- Before every Join component.
- After the input file is read.
- After every 15-20 Component (or) MAX_CORE Parameter is greater than 100%
- None of the above.
17) If I have 10 Psets that can run parallel. What method would you recommend to load the data parallel mode with cost effectiveness :
- Use the Abinitio Plan and schedule the Psets Parallel.
- Run the Psets in serial Mode.
- Run the psets in Background and check the Status with PID value.
- None of the above.
18) What level of skewedness the data should have in a graph
- 0
- 5
- 1
- 99
19) If there is lot of skewness in graph, To reduce the skewedness what is the possible solution that we can have :
- Use Round Robin Component to equally partition the data.
- Use Partition by Key to partition the data.
- Use filter by Expression.
- Use more key value to reduce the skewness in partition by key component.
20) If I have 1 TB of data which mode should I use to insert the data into DATABASE :
- API Mode
- TPT Mode
- MLOAD
- None of the above.
21) If I have a sorted file and there is a requirement to de-partition the data which component should I use for faster performance :
- Gather
- Merge
- Replicate
- None of the above.
22) If I have four Files lets say A, B, C, and D with data volume 50 GB, 40 MB, 10 MB, 2 KB. Which File should I choose as deriving port while doing a Join :
- File A
- File B
- File C
- File D
23) How can I use the 1 TB of Block compressed lookup File for best results and performance :
- Decompress the file and use it as a normal lookup file.
- Use Dynamic lookup Concept.
- Use Join component.
- Can’t say
24) While sending a file to other server securely which method should I use :
- FTP
- SCP
- SFTP
- B and C Both can be used
25) Which of the below operation cannot be possible using REFORMAT and lookup ?
- Inner Join
- Explicit Join
- Full Outer Join
- None of the above.
26) How to run a script ($AI_BIN/send_email_to_support.ksh) after the successful completion of the graph in the end script?
- if [ $? -eq 0 ]; then
- $AI_BIN/send_email_to_support.ksh
- fi
- if [ $mpjret -eq 0 ]; then
- $AI_BIN/send_email_to_support.ksh
- fi
- if [ $$ -eq 0 ]; then
- $AI_BIN/send_email_to_support.ksh
- fi
- Not Possible
27) When using ABLOCAL for unloading in parallel in the complex SQL statements containing grammer, which among the below statements are correct.
- ABLOCAL prevents the INPUT TABLE component from parsing the SQL
- SQL statements will get passed through to the database
- Both A & B
- None of these
28) How to increase the performace of sort
- Convert data from length-prefixed or delimited to fixed-length
- Use multiple sort in sequence.
- Convert data from fixed length to delimited.
- Use complex dml type for key fields.
29) If the flow after the Replicate component needs to be in a different phase, then where to place the checkpoint/phase break.
- Place the checkpoint or phase break after Replicate
- Checkpoint or phase break should not be placed before/after replicate.
- Place the checkpoint or phase break before Replicate
- Placing checkpoint or phase break it has no impact on the performance of graph.
30) Preferable record type for better maintainability ??
- delimited
- fixed-length
- length-prefixed
- Any type
31) Preferable record type for better performance ??
- delimited
- fixed-length
- length-prefixed
- Any type
32) Preferable record type for better flexibility ??
- delimited
- fixed-length
- length-prefixed
- Any type
33) Conditional record formats are extremely useful for parsing heterogeneous data structures, keeping performance in view when should it be used.
- To parse data but not to process it
- Only while process data
- To parse dat and to process the data
- Conditional record formats should not be used
34) For optimal performace while reading, writing, and manipulating which DML types should be used
- delimited
- fixed-length
- length-prefixed
- Any type
35) How to control the amount of memory consumption in a graph
- Breaking the data up into smaller separate datasets
- Reducing the level of data parallelism
- Using THROTTLE component in graph
- Use the AB_NICE configuration variable
36) What is two-stage routing.
- Running the graph in component folding mode implies running it using two-stage routing.
- To save network resources one can mark all-to all flow as using two-stage routing.
- It is layout of network component running on remote system.
- two-stage routing is a conitnuos component.
37) When to use two-stage routing.
- Use two-stage routing if the there are than 30 component in the graph.
- Use two-stage routing there are continuous component used.
- Use two-stage routing if the all-to-all flows in a graph have more than 30 partitions
- It should not be used at all.
38) When is memory mapping of lookup file not possible
- Lookup file is remote.
- Lookup file is indexed
- Lookup file has complex key set
- When size of lookup file is more than 2GB
39) When a lookup plus reformat is better than JOIN
- You need to make the graph easy to read and understand.
- All but one of inputs are small enough to fit into memory.
- One of your non-driving inputs is too large to fit into memory.
- The lookup file may grow significantly over time.
40) When a lookup plus reformat cannot replace a JOIN
- While performing full outer join
- While performing full inner join
- The joining expression is complex and uses several lookup tables.
- The joining expression involves intervals or pattern matching.
41) What is true about Common projects
- Common projects are always created by Ab Initio environment.
- All common projects are public
- Any project included by another project is, by definition, common
- None of the above
42) When to export a parameter to the environment
- Always export a parameter to the environment
- When the graph is deplyed with “dynamic script generation”
- When a parameter is declared as constant.
- It will be referenced indirectly in some other parameter value expression
43) How to determine whether a sandbox is attached to a technical repository project?
- air sandbox show-common
- air sandbox info
- air sandbox find
- air repository show
44) What are .abi-unc files
- It is an alias of .abinitiorc file.
- .abi-unc files are temporary files used by the checkout process.
- This file contains information for unix connectivity.
- No such file is created or exist in Ab Initio.
45) What is the layout of watcher files
- Watchers data is not written anywhere.
- It places in the AB_WORK_DIR.
- It places in the layout of the component upstream of the watcher.
- It places in the layout of the component downstream of the watcher.
46) When database layout is set to default then which working directory does the Co>Operating System use?
- AB_WORK_DIR
- AB_DATA_DIR
- AB_JOB
- AB_WORK
47) What is ABLOCAL and when shall it be used
- When graph needs to be run locally.
- When complex SQL statement contains grammar that cannot be recognized by the Ab Initio parser when unloading in parallel.
- When Co>Operating system and GDE are installed on same system.
- None of the above
48) If we have a source File and a lookup file – both are multifiles and are partitioned on same key, Which lookup function will give the faster performance ??
- Lookup () Function
- Lookup_local () Function
- Lookup and Lookup_local will give the same performance
- None of the above
49) How to improve the performance of a graph?
- Use a limited number of components in a particular phase
- Use optimum value of max core values for sort and join components
- Minimize the number of sort components
- All of the Above
50) Specify the option for command line equivalent while analysing dependency for generic graphs
- expand
- ignore-sql-errors
- link
- none