Biology Reference
In-Depth Information
and “ls” (list, which provides a list of all the fi les and other directories
in the current directory). On almost all computers, fi les are accessed by
means of a hierarchical tree of directories, which can be navigated by
traveling up and down the branches using the “cd” command. In keep-
ing with the tree metaphor, the topmost directory is usually known as
“root” (labeled “/” in Unix). However, almost anything useful in bio-
informatics will require accessing fi les on other computers; these might
be the computers of colleagues sitting next to you, they might be net-
worked hard drives sitting in the corner of the lab or the basement of
the building, or they might be servers at some distant location. In many
cases, the physical location of the machine being accessed makes little
or no difference and is unknown to the user. The most common way
of connecting to another computer is to use “ssh” or “secure shell”—
this is a system of communication that allows you to log into another
computer remotely using a username and password. For example, to
log into a computer called “tulip” at the EBI, I might type “ssh tulip.
ebi.ac.uk.” Tulip will prompt me for a username and password. If I am
authorized to access tulip, the command prompt will reappear—the
screen and the prompt may look exactly the same as if I were using my
own computer, and I can now navigate around the tulip machine using
exactly the same commands.
This sounds straightforward enough. However, access to such vir-
tual spaces is highly regulated. At the EBI, access to all physical spaces
is controlled by RFID cards—when I arrived, I was provided with an
ID on the fi rst day. Access to virtual spaces is limited by username and
password combinations; by contrast, it took over a week to arrange
my access to all the commonly used computers. Moreover, not all com-
puters are directly accessible—sometimes a series of logins is required,
ssh-ing from one's own computer to computer A and then from A to
B. Some computers can be accessed only by programs (those usually
used for intensive calculation), some are dedicated to particular kinds
of data (databases, for instance), some are for everyday use, some for
use by particular users or groups, some for long-term storage, some for
backup, and some for hosting publicly accessible websites. Figure 4.3
gives a sense of the variety of machines involved and the complicated
way in which they can be connected to one another. Bioinformaticians
use metaphors of space in talking about how they move around these
extended networks: tunnels, fi rewalls, routers, shells, and transfers all
suggest a space in which both the user and the data can move around.
As I learned to how to log into various machines and i nd my way
around the network, my questions would invariably be answered with
