Delta-V is effectively just a change of speed so it has dimensions of $LT^{-1}$ e.g. metres per second.
Suppose you have a rocket with mass $m$ that can generate a thrust (i.e. a force) $F$, then by Newton's first law the acceleration is simply:
$$ a = \frac{F}{m} $$
Acceleration is $dv/dt$, so you get the change in velocity simply by integrating the acceleration from the time you start firing your rocket $t_0$ to the time you turn off the booster $t_1$. If you're accelerating in a straight line and if you can ignore the change in mass of the rocket then this gives the equation we all learned in school physics lessons:
$$ \Delta v = a\Delta t = \frac{F}{m} (t_1 - t_0)$$
where $\Delta v$ is simply the change in velocity. For a real rocket the calculation is more complicated for two reasons. Firstly the acceleration usually isn't in a straight line, and secondly the mass of the propellant changes as it burns so $m$ is a function of time not a constant. That's why the Wikipedia article defines it as an integral. The rsulting $\Delta v$ is not a vector because it doesn't have a defined direction, but it's still a speed so you can work out the change in momentum and kinetic energy just by plugging the speed change into the usual equations.
$\Delta v$ is usually given in kilometers per second. According to the Wikipedia article you mention the $\Delta v$ for Mars is 10.7 km/sec rather than the value of 6.2 you mention.
The $\Delta v$ is useful because it tells you how much work you need to get to your target i.e. how powerful your thruster needs to be and how much fuel you need to carry.
The reason that ISP is in units of per second is because it's quoted using a dimensionless unit i.e. the weight of 1kg divided by the weight of 1kg at the earth's surface.
Why deltaV is not a vector?
Take a very simple example where you're moving the rocket between two stationary objects. You start by accelerating the rocket away from the start point, and there is a $\Delta v$ associated with this. As you approach the destination you fire the rockets again to slow down, and again there is a $\Delta v$ associated with the decelleration. If you're a planetary scientist you add the two $\Delta v$s because you want to know how much fuel is required.
However if you treat $\Delta v$ as a vector you find that the initial $\Delta v$ is positive and the final one is negative, so when you do a vector sum of the two you get $\Delta v = 0$. This is unhelpful as it does not mean you've used zero fuel!
Can you elaborate on ISP a bit?
Impulse is force times time, and it's equal to the change of momentum. Specific impulse is impulse per unit mass of fuel i.e. it's the momentum change you get for every unit mass of fuel burned. So a high ISP means you're very fuel efficient i.e. every kg of fuel burned gives you a large change in momentum and therefore velocity.
Suppose you have a specific impulse $I_{sp}$ given in units of seconds, then if you burn a mass $m$ of fuel the impulse, i.e. change of momentum is given by:
$$ I = I_{sp} \space m \space g $$
where $g$ is the acceleration due to gravity at the Earth's surface i.e. 9.81 m/sec.