A question one might ask at a technical interview:

Explain how high volume and a load-balanced environment can affect one’s coding approach.

My answer would be

1. one might resort to caching of data (perhaps with Memcached, perhaps with Memoized).
2. one might cache database handles (with Apache::DBI) or try to pool them
3. one might create static content as much as possible and certainly be sure to have Apache serve all parts of a page which are static such as images.
4. one might replicate the database

1. Given a text file with a full name on each line, print the 10 most common last names in the file… also do this in SQL.
2. what does use strict do
3. what variable controls inheritance in Perl
4. a sub that modifies a hashref - trick question: they made a copy of the hashref in the modifying sub
5. parse a URL
6. simple INNER JOIN
7. composite indexing question
8. how to check if a slave is up
9. hOW TO to see which processes heavy
10. how to see disk activity
11. How to see disk usage?
12. how to check if apache is up?
13. Design a couple of pages for a yearbook website

MySQL in-depth: Architecture Type Questions

1. What is the lost update problem and how does MySQL solve it?
2. What is a deadlock? Does it happen with a SQL approach to the lost update problem?
3. How can you solve it without logical units of work?
4. Discuss relation normalization and denormalization
5. What is a primary key and for InnoDB tables discuss how the physical ordering of data relates to the primary key

Filed under: Perl Software Design Blog » competency-tests » mysql » arch-level

// show without comments // write a comment // Trackback URL

Garth’s test

sort

iterator

Artur’s test

The monk climbs the mountain from 6am to 6pm. And the next day while he is descending from 6am to 6pm, a priest climbs. The priest gets to the top at 6pm, which is the same time that the monk gets to the bottom.

Is there a time that they are ever at the same height?

I solved this equationally at shopzilla somehow. I solved it graphically at shopzilla. The next day, I realized, that the picture of someone going down the other side was a red herring: if I had only reflected the downward trip on the upward trip, I would have noticed that they would cross paths!

Filed under: Perl Software Design Blog » competency-tests » sixapart » artur

// show without comments // write a comment // Trackback URL

My second technical test at IMDB

1. If a SELECT is running too slowly how would you debug it?
2. why not index a bunch of individual fields in a database
3. Class::DBI iterators - how are they implemented? Are they efficient?
4. how does CDBI differ from other DBIx::*
5. least favorite Perl feature
6. what are you going to learn next about Perl?

Filed under: Perl Software Design Blog » competency-tests » imdb » second

// show without comments // write a comment // Trackback URL

Perl competency test (imdb)

I took a small but good competency test when interviewing at imdb.

The cvs module name is imdb-perl-test and it should be available via viewcvs in 3 hours.

Ah here it is

Filed under: Perl Software Design Blog » competency-tests » imdb » first

// show without comments // write a comment // Trackback URL

Perl SQL competency test (sdc)

I took a small but good competency test when interviewing at a small but good company here in Los Angeles :). It only had one Perl question and one SQL question and I was lucky to get the SQL question right.

The cvs module name is perl-sql-competency-test and it should be available via viewcvs in 3 hours.

Filed under: Perl Software Design Blog » competency-tests » small-perl-sql

// show without comments // write a comment // Trackback URL