In the context of IT infrastructure operation, a methodology has been proposed that describe both operation instructions and commands in executable runbooks using Jupyter Notebook to achieve highly reproducible operation. However, to apply this methodology to remote infrastructure operation, operators are forced to rewrite too many commands in typical operation manuals. To address this issue, we developed a Jupyter extension to both solve this problem and improve the usability of executable runbooks. We then confirmed the effectiveness of this extension in a user study.

Adaptive-ness of multicast stream is a strongly desired for P2P multicast to be accepted broadly on the Internet. One way to achieve this is use the layered multicast technique. But it was pointed out that there are obstacles to apply it on the Internet. In this paper, we apply TFRC (TCP Friendly Rate Control) scheme onto XCAST, while TFRC is considered basically for unicast. As a proof of concept, we have implemented TFRC with XCAST onto DVTS and we were able to confirm that they achieve necessary characteristics for P2P multicast such as stability, flexibility, fairness and smoothness. These characteristics were independently observed for each sub-path of XCAST tree.

This paper proposes a new type of layered video coding especially for the use of monitoring a river or a water channel. A sensor node of the system decomposes a video signal into some components and produces a bit stream which is functionally separated into three layers. The first layer contains the minimum components effective for detecting water level. The second layer contains signals for thumb-nail video browsing. Each of them is transmitted at very low bit rate for regular monitoring. The third layer contains additional data for decoding the original video signal. It is transmitted in case of necessity. The original video signal is decomposed into band signals as the components by the Haar transform in a sensor node. Experimental results show which band signals should be included into the first layer considering both of water level detection performance and data size to be transmitted.

CrispORB is a high performance CORBA that enables low-latency request delivery among cluster nodes connected with the system area network (SAN) and provides a communication infrastructure for large-scale and high-volume cluster system construction with distributed object-oriented components. Compared with standard IIOP ORBs, CrispORB is be able to reduce the protocol overhead using a system area network with intelligent functions for efficient data transfer. CrispORB preserves language mapping specified by COBRA/OMG standards so that component programs written for conventional CORBA could be used without rewriting. CrispORB can manage both SAN and TCP/IP simultaneously, which means the cluster node can communicate with clients via the Internet and with other nodes on the cluster using SAN. Components need not be aware of lower connections because CrispORB selects the transport between SAN and TCP/IP automatically. We selected the VI Architecture to use SAN's communication functions, which are proposed as an industrial standard API for SAN. CrispORB will be portable onto various types of SANs. We have integrated VIA buffer management and CDR (Common Data Representation) format processing to reduce data movement cost. CrispORB was developed based on INTERSTAGE/ObjectDirector (Fujitsu'S CORBA product) and Synfinity-O (Fujitsu's SAN).