Precision studies of the NNLO DGLAP evolution at the LHC with Candia

We summarize the theoretical approach to the solution of the NNLO DGLAP equations using methods based on the logarithmic expansions in x-space and their implementation into the C program Candia 1.0.1 We present the various options implemented in the program and discuss the different solutions. The user can choose the order of the evolution, the type of the solution, which can be either exact or truncated, and the evolution either with a fixed or a varying flavor number, implemented in the varying-flavor-number scheme (VFNS). The renormalization and factorization scale dependencies are treated separately. In the non-singlet sector the program implements an exact NNLO solution.Program summaryProgram title: CANDIACatalogue identifier: AEBK_v1_0Program summary URL:http://cpc.cs.qub.ac.uk/summaries/AEBK_v1_0.htmlProgram obtainable from: CPC Program Library, Queen's University, Belfast, N. IrelandLicensing provisions: Standard CPC licence, http://cpc.cs.qub.ac.uk/licence/licence.htmlNo. of lines in distributed program, including test data, etc.: 101 376No. of bytes in distributed program, including test data, etc.: 5 865 234Distribution format: tar.gzProgramming language: C and FortranComputer: AllOperating system: LinuxRAM: In the given examples, it ranges from 4 to 490 MBClassification: 11.1, 11.5Nature of problem: The program provided here solves the DGLAP evolution equations for the parton distribution functions up to NNLO.Solution method: The algorithm implemented is based on the theory of the logarithmic expansions in Bjorken x-space.Additional comments: To be sure of getting the latest version of the program, the authors suggest downloading the code from their official CANDIA website (http://www.le.infn.it/candia).Running time: In the given examples, it ranges from 1 to 40 minutes. The jobs have been executed on an Intel Core 2 Duo T7250 CPU at 2 GHz with a 64 bit Linux kernel. The test run script included in the package contains 5 sample runs and may take a number of hours to process, depending on the speed of the processor used and the size of the available RAM.