Sign In

Communications of the ACM

Kode Vicious

A Lesson in Resource Management


View as: Print Mobile App ACM Digital Library Full Text (PDF) In the Digital Edition Share: Send by email Share on reddit Share on StumbleUpon Share on Hacker News Share on Tweeter Share on Facebook
George V. Neville-Neil

George V. Neville-Neil

back to top  Dear KV

I have been reworking a device driver for a high-end, high-performance networking card and I have a resource allocation problem. The devices I am working with have several network ports, but these are not always in use; in fact, many of our customers use only one of the four available ports. It would greatly simplify the logic in my driver if I could allocate the resources for all the portsno matter how many there arewhen the device driver is first loaded into the system, instead of dealing with allocation whenever an administrator brings up an interface.

I should point out that this device has a good deal of complexity and the resource allocation is not as simple as a quick malloc of memory and pointer jigglinga lot of moving parts are inside this thing.

We are not talking about a huge amount of memory by modern standardsperhaps a megabyte per portbut it still bothers me to waste memory, or really any resource, if it is not going to get used. I am old enough to remember eight-bit computers with 64 kilobytes of RAM, and programming those gave me a strong internal incentive never to waste a byte, let alone a megabyte. When is it OK to allocate memory that might never be used, even if this might reduce the complexity of my code?

Fearful of Footprints

Back to Top

Dear Footprints

The answer to your question is easy. It is sometimes OK to allocate memory that might never be used, and it is sometimes not OK to allocate the same memory. Ah, are those the screams of a programmer without a black-and-white, true-or-false answer to the question that I hear? Delightful!

Software engineering, much to your and my chagrin, is the study of trade-offs. Time vs. complexity, expediency vs. qualitythese are the choices we deal with every day. It is important for engineers to revisit their assumptions periodically, perhaps every year or two, as the systems we work on change under us quite quickly.

Programmers who are paying attention to the systems they useand I know that each and every one of my readers is paying attentionhave seen these systems change dramatically over the past five years, just as they had the five years before that, and so on, back to the first computers. While processor frequency scaling may have paused for the moment (and we will see how long that moment lasts), the size of memory has continued to grow. It is not uncommon to see single servers with 64 and 128 megabytes of RAM, and this explosion of available memory has led to some very poor programming practices.

Blindly wasting resources, such as memory, really is foolish, but in this case it is not an engineering trade-off, it is an example of a programmer who is too far from their machine trying to "just make it work." That is not programming, that is just typing. Software engineers and programmers worth their expensive chairs and high salaries know they do not want to waste resources, so they try to figure out what the best- and worst-case scenarios are and how they will affect the other possible users of the system. Users in most cases are now just other programs, rather than other people, but we all know what happens to a system when it starts to swap things out of memory onto secondary storage. That's right, your DevOps people call you screaming at 3 A.M. Screaming people are never that much fun, except at a concert.

You mentioned this software is for a "high-performance" device, and if by that you mean it goes in a typical 64-bit server-class machine, then no one is really going to notice a megabyte, or four, or even eight. A high-end server-class machine is unlikely to have less than four gigabytes of RAM. Even if you allocate four megabytes at system startup time, that is one-tenth of 1% of the available RAM. People writing in Java will suck down far more than that just starting their threads. Are you really going to worry about less than one-tenth of a percent of memory?

If you had told me that this driver was for some limited-memory-size embedded device, I would give other advice, since that system might not have 4GB of RAM; but then again, given what most phones and tablets have in them now, it might.

People are often right when they say, "Waste not, want not," but it is also important to take your moderation in moderation.

KV

Back to Top

Dear KV,

I am converting a Web application to run on a mobile platform. While the application does not handle banking information or anything with crazy security like that, it does still require the user to type a password. Our password requirements are not too strict, though we do have a minimum size of eight characters and require one uppercase letter and one nonalphanumeric character. Because on-screen typing is so inaccurate, our product development folks want us to relax our password requirements even more, allowing the user to have a four-character, all-lowercase password, which they call a mobile PIN code. The mobile PIN would work only from the mobile app and not on the Web. I have tried to explain to the product development group that four characters simply are not enough, but maybe if we are restricting this to the mobile app, it will be OK. What do you think?

Pinned Down

Back to Top

Dear Pinned

I have been wondering when someone would write a letter like this. Having typed on a variety of tablets in the past few years, I knew it was only a matter of time before some marketing or product-type person would ask an engineer to dumb down security for convenience. I am glad you pointed out your application does not relate to banking, as I would have had to write back immediately and ask, "Which bank?!", and that could only lead to trouble.

As a quick aside, I do find it interesting that the world thinks only of banks when they think of security. While it would be very bad for someone to transfer all the money out of your bank account into their own, the fact of the matter is that a lot of really bad things can happen online that do not relate directly to cash flow. Weak passwords can leave users open to identity theft, stalkers, kidnappers, and their exes. It is up to you to think about which of those is better or worse than losing some cash. I would take losing money over being stalked by my exes.


The problem with a four-character password is that it is too short and easy to guess.


The problem with a four-character passwordas you point outis that it is too short and easy to guess. A PIN with a card, such as a card used at an ATM, is employed because it requires physical possession of something, the card, to be effective. Schemes for protecting users online, such as those involving passwords, depend on the user presenting a combination of something they have, something they are, or something they know. A password is an example of the latter. The schemes can be mixed and matched, such as the PIN and a physical card, where the user has something and knows something.

The problem with current mobile devices is that they are much more limited in their ability to handle user input than, say, a computer with a keyboard. On-screen keyboards are not very good, which is probably why someone at Google thought to use a pattern for device unlocking on the Android. I find that system a bit silly and easy to shoulder surf, but it is good to see someone trying to do something differently.

It would be nice to pretend that the device itself would be proof of something the user has, but since the point of the password is to prevent a malicious party from accessing the user's data after the device has been lost or stolen, it needs to be sufficiently strong to deter an attacker. If you are unable to fight back on password complexity, it is time to break out the lockout option. A four-digit pin code is mostly a problem if you allow the attacker a large number of attempts to guess the PIN. If, after three tries, you lock out the user for five minutes, and then let the user try again, it is going to take a long time for the attacker to try enough PINs to guess the right onethat is, if the user has not picked a common PIN, such as 1234 or 2580 (an exercise for readers is figuring out why that second code is so common; for a more academic study of the problem of PINs see the paper, "A birthday present every eleven wallets? The security of customer-chosen banking PINs," by Joseph Bonneau et al from the Computer Laboratory of the University of Cambridge at http://www.jbonneau.com/doc/BPA12-FC-banking_pin_security.pdf).

Being the kind of person I am, I also like the idea of three failed tries causing the device not to work at all, including erasing all local data and then requiring the user to go through a recovery flow that does not involve the device. If your application is used by people who are often mentally impaired in some way (and no, I do not mean they are upper management, but I do mean when it is used for social networking), then an annoying recovery system is not going to get past your product development folks. Too many people want to upload pictures of their friends in compromising positions, and that, alas, requires compromising on security.

KV

q stamp of ACM QueueRelated articles
on queue.acm.org

Power-Efficient Software
Eric Saxe
http://queue.acm.org/detail.cfm?id=1698225

Network Virtualization: Breaking the Performance Barrier
Scot Rixner
http://queue.acm.org/detail.cfm?id=1348592

Kode Vicious: The Doctor Is In
George Neville-Neil
http://queue.acm.org/detail.cfm?id=1105672

Back to Top

Author

George V. Neville-Neil (kv@acm.org) is the proprietor of Neville-Neil Consulting and co-chair of the ACM Queue editorial board. He works on networking and operating systems code for fun and profit, teaches courses on various programming-related subjects, and encourages your comments, quips, and code snips pertaining to his Communications column.


Copyright held by Owner/Author(s).

The Digital Library is published by the Association for Computing Machinery. Copyright © 2013 ACM, Inc.