The Message Passing Interface (MPI) has become a standard for message passing parallel applications. This report first introduces the underlying paradigm, message passing, and explores some of the challenges explicit message passing poses for developing parallel programs. We then take a closer look at the MPI standardization effort, its goals, and its results to see what features the current ve...

Open MPI was initially designed to support a wide variety of high-performance networks and network programming interfaces. Recently, Open MPI was enhanced to support networks that have full support for MPI matching semantics. Previous Open MPI efforts focused on networks that require the MPI library to manage message matching, which is sub-optimal for some networks that inherently support match...

DEVICE INTERFACE ABSTRACT DEVICE INTERFACE Figure 14 Digital MPI Architecture Table 5 MPI Latency Comparison MPI Implementation Transport Platform Latency MPICH 1.0.10 Sockets FDDI DEC 3000/800 350 ms MPICH 1.0.10 Shared Memory AlphaServer 2100 4/233 30 ms Digital MPI V1.0 MEMORY CHANNEL 1.0 AlphaServer 2100 4/233 16 ms Digital MPI V1.0 MEMORY CHANNEL 1.5 AlphaServer 4100 5/300 6.9 ms Digital M...

OBJECTIVE The diagnostic yield of combined cardiopulmonary exercise testing (CPET) and myocardial perfusion imaging (MPI) in patients referred for stress testing has received limited study. METHODS We evaluated consecutive patients who underwent combined CPET-MPI at a single tertiary referral center between 2011 and 2015. An abnormal CPET was defined as any of the following: reduced oxygen co...

In scientific applications, netCDF [1] was developed to support a view of data as a collection of self-describing, portable, and array-oriented objects that can be accessed through a simple interface. Its parallel I/O interface named parallel netCDF (hereafter PnetCDF) was developed with the help of an MPI-I/O library [2] such as ROMIO [3], and the PnetCDF succeeded in scientific computation [4...

As parallel systems are commonly being built out of increasingly large multicore chips, application programmers are exploring the use of hybrid programming models combining MPI across nodes and multithreading within a node. Many MPI implementations, however, are just starting to support multithreaded MPI communication, often focussing on correctness first and performance later. As a result, bot...

The Message Passing Interface (MPI) is the prevalent programming model used on today’s supercomputers. Therefore, MPI library developers are looking for the best possible performance (shortest run-time) of individual MPI functions across many different supercomputer architectures. Several MPI benchmark suites have been developed to assess the performance of MPI implementations. Unfortunately, t...

The Message Passing Interface (MPI) 3.0 standard includes a significant revision to MPI’s remote memory access (RMA) interface, which provides support for one-sided communication. MPI-3 RMA is expected to greatly enhance the usability and performance of MPI RMA. We present the first complete implementation of MPI-3 RMA and document implementation techniques and performance optimization opportun...

The number of multithreaded Message Passing Interface (MPI) implementations and applications is increasing rapidly. We discuss how multithreaded applications can receive messages of unknown size. As is well known, combining MPI Probe/MPI Recv is not threadsafe, but many assume that trivial workarounds exist. We discuss those workarounds and show how they fail in practice by either limiting the ...

OBJECTIVES To evaluate the interobserver reliability of echocardiographic findings of right ventricle (RV) dysfunction for prognosticating normotensive patients with pulmonary embolism (PE). METHODS A central panel of cardiologists evaluated echocardiographic studies of 75 patients included in the PROTECT study for the following signs: RV diameter, RV/left ventricular (LV) diameter ratio, hyp...