Network monitoring and measurement is an invaluable tool for comprehending, analyzing, managing, and optimizing performance and security of networked systems. Network monitoring architectures can take the form of local or distributed deployments of sensors. Local deployments can be very precise and efficient because they benefit from fast links to the central monitoring station, but their scope can be limited to local or small-scale networks. Distributed monitoring infrastructures give a much broader view of the network state, but have the disadvantage that the amount of information they can push back to the central monitoring station is limited by the capacity of the links. In this paper we investigate the effects of compression on network monitoring data streams that are transmitted from distributed network sensors back to a central infrastructure. Our analysis shows that we can achieve very high compression rates, which means we remove much of the capacity overheads when transmitting sensor data back to the central monitors, while incurring only minimal delay in transmission of monitoring information. Our scheme also has the additional benefit of decreased CPU load at the monitoring sensor due to the aggregation of data which reduces the number of network messages.