Quantum information science and atom optics are among the most active fields in modern physics. In recent years, many theoretical efforts have been made to combine these two fields. Recent experimental progresses have shown the in-principle possibility to perform scalable quantum information processing (QIP) with linear optics and atomic ensembles. The main purpose of the present project is to use atomic qubits as quantum memory and exploit photonic qubits for information transfer and processing to achieve efficient linear optics QIP. On the one hand, utilizing the interaction between laser pulses and atomic ensembles we will experimentally investigate the potentials of atomic ensembles in the gas phase to build quantum repeaters for long-distance quantum communication, that is, to develop a new technological solution for quantum repeaters making use of the effective qubit-type entanglement of two cold atomic ensembles by a projective measurement of individual photons by spontaneous Raman processes. On this basis, we will further investigate the advantages of cold atoms in an optical trap to enhance the coherence time of atomic qubits beyond the threshold for scalable realization of quantum repeaters. Moreover, building on our long experience in research on multi-photon entanglement, we also plan to perform a number of significant experiments in the field of QIP with particular emphasis on fault-tolerant quantum computation, photon-loss-tolerant quantum computation and cluster-state based quantum simulation. Finally, by combining the techniques developed in the above quantum memory and multi-photon interference experiments, we will further experimentally investigate the possibility to achieve quantum teleportation between photonic and atomic qubits, quantum teleportation between remote atomic qubits and efficient entanglement generation via classical feed-forward. The techniques that will be developed in the present project will lay the basis for future large scale