Heterogeneous Spacecraft Networks (HSNs) are network environments in which spacecraft from different missions and institutions can communicate with each other at low cost and with low impact on overall system resources. The Mission Design Center (MDC) at NASA Ames Research Center has been studying solutions for low cost multi spacecraft systems for a number of years. One may now build on the idea to interconnect clusters of spacecraft with each other to have them act as mobile nodes belonging to the same collaborative mission. Recent progress in small satellite technology is significant, and one of the advantages of small satellites lies precisely in the large quantity of spacecraft that can be produced at accessible costs. It follows naturally that small satellites are an interesting candidate platform for development and demonstration of the HSN concept. This paper is the second in a series of three companion papers. The general concept of operations for HSNs in LEO and a number of future applications are proposed in the first paper [6], while enabling technology such as devices and lower layer protocols are discussed in paper three [7]. In this paper, we pick up the scenario of a low-cost and multi-institutional network of Earth Observation (EO) missions in LEO and conduct network performance analysis using the AGI System Tool Kit (STK) and the open-source Network Simulator (NS-3). A multi spacecraft network consolidates the individual capabilities of each spacecraft from different institutions by combining benefits of both frequent revisit and concentrated observation. Complementary and correlated data could be collected simultaneously from a large set of distributed spacecraft utilizing HSN capability. In this specific configuration, communication distance between spacecraft, related delays and error rate are the major factors in network performance.Also, average duration of communication opportunities between spacecraft is usually very limited. Thus, it is important to simulate orbital dynamics, link margins, and protocols simultaneously to analyze network performances. In this paper, we compare some existing protocols to obtain a measure for the practical performance of the candidate network. We focus on best-effort data delivery, an approach necessitated by the severe constraints on communications resulting from low-cost and low system resource small spacecraft. In the application layer, we show that packet size and data rate of a source node also affect overall performance of the network. We present the resulting figures of merit from our simulations. The paper concludes with a summary of the simulation results.