Fun fact relating to fusion reactors: They can easily (at least until Greg's newest nerf to them, which probably won't be coming to FTB until 1.5, in which the multiblock structure also became MUCH larger) be automated to run deuterium and tritium (using only 7 industrial electrolyzers and 18 centrifuges)- but sustainably running one on deutirium and helium-3 requires EIGHT reactors running tritium, as well as quite a few centrifuges (I assume, I haven't done the actual math there yet).
Why, you ask? Well, the only ways to get helium-3 are from endstone and centrifuging 16 helium-4. The only ways to get regular helium are from endstone again, glowstone, and nuclear fusion. As it turns out, both aforementioned reactions produce one helium cell per cycle- so you need sixteen fusion cycles to end during the time it takes your helium-3 cycle to finish. The helium-3 fusion takes twice as long as tritium fusion, so you only need eight tritium reactors.
Also, while you're at it (if you've already built nine fusionreactors), you might as well create a tenth one to dedicate to making iridium. I remember that there are two reactions that require a constant supply of power, but I don't remember the recipes. I think one is lithium + tungsten = iridium, and I've found that clay is a good source of lithium (5 UUM = 48 clay balls, if I remember correctly, and electrolyzing however many clay dust takes only 10 seconds- plus you also get sodium, silicon, and aluminum), and tungsten comes from centrifuging lava (as well as copper, tin, and electrum (centrifuge again, I think, for gold and silver)).
Also, it seems that emeralds are a good source of berylium. Much better than enderpearls, if you want to make your reactors out of nothing but UUM (which can certainly be done).
Also, DoctorOr, what UUM cost balance are you using? Vanilla IC2's "cheap" recipe or GregTech's default 100 times more expensive recipe?