Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “TCP”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 199 records · Page 11
All sourcesNASA NTRSDOE OSTI
DOE OSTI2015

Volumetric Modulated Arc Therapy Planning for Primary Prostate Cancer With Selective Intraprostatic Boost Determined by {sup 18}F-Choline PET/CT

Purpose: This study evaluated expected tumor control and normal tissue toxicity for prostate volumetric modulated arc therapy (VMAT) with and without radiation boosts to an intraprostatically dominant lesion (IDL), defined by {sup 18}F-choline positron emission tomography/computed tomography (PET/CT). Methods and Materials: Thirty patients with localized prostate cancer underwent {sup 18}F-choline PET/CT before treatment. Two VMAT plans, plan{sub 79} {sub Gy} and plan{sub 100-105} {sub Gy}, were compared for each patient. The whole-prostate planning target volume (PTV{sub prostate}) prescription was 79 Gy in both plans, but plan{sub 100-105} {sub Gy} added simultaneous boost doses of 100 Gy and 105 Gy to the IDL, defined by 60% and 70% of maximum prostatic uptake on {sup 18}F-choline PET (IDL{sub suv60%} and IDL{sub suv70%}, respectively, with IDL{sub suv70%} nested inside IDL{sub suv60%} to potentially enhance tumor specificity of the maximum point dose). Plan evaluations included histopathological correspondence, isodose distributions, dose-volume histograms, tumor control probability (TCP), and normal tissue complication probability (NTCP). Results: Planning objectives and dose constraints proved feasible in 30 of 30 cases. Prostate sextant histopathology was available for 28 cases, confirming that IDL{sub suv60%} adequately covered all tumor-bearing prostate sextants in 27 cases and provided partial coverage in 1 case. Plan{sub 100-105} {sub Gy} had significantly higher TCP than plan{sub 79} {sub Gy} across all prostate regions for α/β ratios ranging from 1.5 Gy to 10 Gy (P<.001 for each case). There were no significant differences in bladder and femoral head NTCP between plans and slightly lower rectal NTCP (endpoint: grade ≥ 2 late toxicity or rectal bleeding) was found for plan{sub 100-105} {sub Gy}. Conclusions: VMAT can potentially increase the likelihood of tumor control in primary prostate cancer while observing normal tissue tolerances through simultaneous delivery of a steep radiation boost to a {sup 18}F-choline PET-defined IDL.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2016

Predicting the 5-Year Risk of Biochemical Relapse After Postprostatectomy Radiation Therapy in ≥PT2, pN0 Patients With a Comprehensive Tumor Control Probability Model

Purpose: To fit the individual biochemical recurrence-free survival (bRFS) data from patients treated with postprostatectomy radiation therapy (RT) with a comprehensive tumor control probability (TCP) model. Methods and Materials: Considering pre-RT prostate-specific antigen (PSA) as a surrogate of the number of clonogens, bRFS may be expressed as a function of dose-per-fraction–dependent radiosensitivity (α{sub eff}), the number of clonogens for pre-RT PSA = 1 ng/mL (C), and the fraction of patients who relapse because of clonogens outside the treated volume (K), assumed to depend (linearly or exponentially) on pre-RT PSA and Gleason score (GS). Data from 894 node-negative, ≥pT2, pN0 hormone-naive patients treated with adjuvant (n=331) or salvage (n=563) intent were available: 5-year bRFS data were fitted grouping patients according to GS (<7:392, =7:383, >7:119). Results: The median follow-up time, pre-RT PSA, and dose were 72 months, 0.25 ng/mL, and 66.6 Gy (range 59.4-77.4 Gy), respectively. The best-fit values were 0.23 to 0.26 Gy{sup −1} and 10{sup 7} for α{sub eff} and C for the model considering a linear dependence between K and PSA. Calibration plots showed good agreement between expected and observed incidences (slope: 0.90-0.93) and moderately high discriminative power (area under the curve [AUC]: 0.68-0.69). Cross-validation showed satisfactory results (average AUCs in the training/validation groups: 0.66-0.70). The resulting dose-effect curves strongly depend on pre-RT PSA and GS. bRFS rapidly decreases with PSA: the maximum obtainable bRFS (defined as 95% of the maximum) declined by about 2.7% and 4.5% for each increment of 0.1 ng/mL for GS <7 and ≥7, respectively. Conclusions: Individual data were fitted by a TCP model, and the resulting best-fit parameters were radiobiologically consistent. The model suggests that relapses frequently result from clonogens outside the irradiated volume, supporting the choice of lymph-node irradiation, systemic therapy, or both for specific subgroups (GS <7: PSA >0.8-1.0 ng/mL; GS ≥7: PSA >0.3 ng/mL). Early RT should be preferred over delayed RT; the detrimental effect of PSA increase can never be fully compensated by increasing the dose, especially for patients with GS ≥7.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2012

Influence of composition on microstructural parameters of single crystal nickel-base superalloys

Fourteen nickel-base superalloy single crystals containing a range of chromium (Cr), cobalt (Co), molybdenum (Mo), and rhenium (Re) levels, and fixed amounts of aluminum (Al) and tantalum (Ta), were examined to determine the effect of bulk composition on basic microstructural parameters, including {gamma} Prime solvus, {gamma} Prime volume fraction, topologically close-packed (TCP) phases, {gamma} and {gamma} Prime phase chemistries, and {gamma}-{gamma} Prime lattice mismatch. Regression models describing the influence of bulk alloy composition on each of the microstructural parameters were developed and compared to predictions by a commercially-available software tool that used computational thermodynamics. Co produced the largest change in {gamma} Prime solvus over the wide compositional range explored and Mo produced the biggest effect on the {gamma} lattice parameter over its range, although Re had a very potent influence on all microstructural parameters investigated. Changing the Cr, Co, Mo, and Re contents in the bulk alloy had an impact on their concentrations in the {gamma} matrix and to a smaller extent in the {gamma} Prime phase. The software tool under-predicted {gamma} Prime solvus temperatures and {gamma} Prime volume fractions, and over-predicted TCP phase volume fractions at 982 Degree-Sign C. However, the statistical regression models provided excellent estimations of the microstructural parameters and demonstrated the usefulness of such formulas. - Highlights: Black-Right-Pointing-Pointer Effects of Cr, Co, Mo, and Re on microstructure in new low density superalloys Black-Right-Pointing-Pointer Co produced a large change in {gamma} Prime solvus; Mo had a large effect on lattice mismatch. Black-Right-Pointing-Pointer Re exhibited very potent influence on all microstructural parameters was investigated. Black-Right-Pointing-Pointer {gamma} and {gamma} Prime phase chemistries both varied with temperature and alloy composition. Black-Right-Pointing-Pointer Computational thermodynamic modeling tool did not accurately predict microstructure.

36 MATERIALS SCIENCE↗
DOE OSTI2017

Microstructural characteristics of σ phase and P phase in Ru-containing single crystal superalloys

Microstructural instability caused by topologically close-packed (TCP) phase precipitation restricts the useful compositional range of advanced Ni-base single crystal superalloys in industrial applications. Limited systematic investigations of TCP formers (Cr and Mo) additions on microstructural evolution of both the σ phase and the P phase in Ru-containing single crystal superalloys have been reported. In this study, the microstructural characteristics of σ phase and P phase were investigated in three Ru-containing superalloys with different levels of Cr and Mo additions at 950 °C and 1100 °C by using phase extraction, X-ray diffraction, scanning electron microscope and high resolution transmission electron microscopy. The experimental results indicated that the high level additions of Cr and Mo promoted the formation of σ phase and P phase, respectively. The amount of σ phase was much higher than that of P phase after long term exposure at 950 °C and 1100 °C. The sheet-like σ phase existed in the alloy with higher Cr addition after thermal exposure at 950 °C and 1100 °C for 1000 h, while the needle-like P phase precipitated in high Mo content alloy after thermal exposure at 1100 °C for 1000 h and the intergrowth of σ phase and P phase was observed after thermal exposure at 950 °C for 500 h. Both the σ phase and P phase were enriched in Re, W, Cr and Mo, but the σ phase contained more Re and Cr while the P phase contained more Mo and Ni, and Ru was found in both phases. The nucleation of σ phase was much easier than P phase due to the more ledge steps in the interfacial structure between σ phase and matrix, as well as the higher partitioning ratios of Re, Cr and Mo. This study is helpful to understand the microstructural evolution of σ phase and P phase, and to optimize the alloy design in Ru-containing superalloys. - Highlights: •Microstructures of σ phase and P phase were characterized in detail. •Cr and Mo influenced the precipitation of σ phase and P phase, respectively. •Partitioning ratios and interfacial relationship decided precipitation behaviors.

36 MATERIALS SCIENCE↗
DOE OSTI2010

SORS: A New Software for the Simulation of Radiotherapy Schedule

We present a software for choosing the best radiotherapy treatment schedule for head and neck cancers as a beginning radiotherapy plan or a temporarily interrupted plan. Its application occurs according to two modalities: the first adopts the best estimates for model parameters; the second takes into account the parameters' uncertainty too. In both cases, the choice becomes the schedule with the highest uncomplicated tumor control probability (UTCP). In the UTCP valuation, the normal tissue complication probability (NTCP) of each organ is related to the gravity of its possible late injury. For NTCP calculation, it has been adopted the empirical LKB (Lyman-Kutcher-Burman) model corrected for dose/fraction via linear-quadratic model and the incomplete repair effect. The tumor control probability (TCP) model is Poisson based and contains corrections for dose/fraction and regrowth effect; optionally, it can be accounted for the incomplete repair effect as well. At the end of processing, a detailed file with all informations about UTCP, TCP and single organ NTCP is furnished for every examined schedule. Moreover, a useful 3-D graphic representation of the schedule's UTCP is available, allowing the physician to easily understand the schedules with the highest radiotherapeutic efficacy. The open source characteristic allows the program to adapt to the individual clinical case as well as to be a valid support in radiobiological research.

61 RADIATION PROTECTION AND DOSIMETRY↗
DOE OSTI2011

Tissue Heterogeneity in IMRT Dose Calculation for Lung Cancer

The aim of this study was to evaluate the differences in accuracy of dose calculation between 3 commonly used algorithms, the Pencil Beam algorithm (PB), the Anisotropic Analytical Algorithm (AAA), and the Collapsed Cone Convolution Superposition (CCCS) for intensity-modulated radiation therapy (IMRT). The 2D dose distributions obtained with the 3 algorithms were compared on each CT slice pixel by pixel, using the MATLAB code (The MathWorks, Natick, MA) and the agreement was assessed with the {gamma} function. The effect of the differences on dose-volume histograms (DVHs), tumor control, and normal tissue complication probability (TCP and NTCP) were also evaluated, and its significance was quantified by using a nonparametric test. In general PB generates regions of over-dosage both in the lung and in the tumor area. These differences are not always in DVH of the lung, although the Wilcoxon test indicated significant differences in 2 of 4 patients. Disagreement in the lung region was also found when the {Gamma} analysis was performed. The effect on TCP is less important than for NTCP because of the slope of the curve at the level of the dose of interest. The effect of dose calculation inaccuracy is patient-dependent and strongly related to beam geometry and to the localization of the tumor. When multiple intensity-modulated beams are used, the effect of the presence of the heterogeneity on dose distribution may not always be easily predictable.

61 RADIATION PROTECTION AND DOSIMETRY↗
DOE OSTI2015

Postoperative radiotherapy following mastectomy for patients with left-sided breast cancer: A comparative dosimetric study

The purposes of this article were to compare the biophysical dosimetry for postmastectomy left-sided breast cancer using 4 different radiotherapy (RT) techniques. In total, 30 patients with left-sided breast cancer were randomly selected for this treatment planning study. They were planned using 4 RT techniques, including the following: (1) 3-dimensional conventional tangential fields (TFs), (2) tangential intensity-modulated therapy (T-IMRT), (3) 4 fields IMRT (4F-IMRT), and (4) single arc volumetric-modulated arc therapy (S-VMAT). The planning target volume (PTV) dose was prescribed 50 Gy, the comparison of target dose distribution, conformity index, homogeneity index, dose to organs at risk (OARs), tumor control probability (TCP), normal tissue complication probability (NTCP), and number of monitor units (MUs) between 4 plans were investigated for their biophysical dosimetric difference. The target conformity and homogeneity of S-VMAT were better than the other 3 kinds of plans, but increased the volume of OARs receiving low dose (V{sub 5}). TCP of PTV and NTCP of the left lung showed no statistically significant difference in 4 plans. 4F-IMRT plan was superior in terms of target coverage and protection of OARs and demonstrated significant advantages in decreasing the NTCP of heart by 0.07, 0.03, and 0.05 compared with TFs, T-IMRT, and S-VMAT plan. Compared with other 3 plans, TFs reduced the average number of MUs. Of the 4 techniques studied, this analysis supports 4F-IMRT as the most appropriate balance of target coverage and normal tissue sparing.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2017

Bladder radiotherapy treatment: A retrospective comparison of 3-dimensional conformal radiotherapy, intensity-modulated radiation therapy, and volumetric-modulated arc therapy plans

To examine tumor's and organ's response when different radiotherapy plan techniques are used. Ten patients with confirmed bladder tumors were first treated using 3-dimensional conformal radiotherapy (3DCRT) and subsequently the original plans were re-optimized using the intensity-modulated radiation treatment (IMRT) and volumetric-modulated arc therapy (VMAT)-techniques. Targets coverage in terms of conformity and homogeneity index, TCP, and organs' dose limits, including integral dose analysis were evaluated. In addition, MUs and treatment delivery times were compared. Better minimum target coverage (1.3%) was observed in VMAT plans when compared to 3DCRT and IMRT ones confirmed by a statistically significant conformity index (CI) results. Large differences were observed among techniques in integral dose results of the femoral heads. Even if no statistically significant differences were reported in rectum and tissue, a large amount of energy deposition was observed in 3DCRT plans. In any case, VMAT plans provided better organs and tissue sparing confirmed also by the normal tissue complication probability (NTCP) analysis as well as a better tumor control probability (TCP) result. Our analysis showed better overall results in planning using VMAT techniques. Furthermore, a total time reduction in treatment observed among techniques including gantry and collimator rotation could encourage using the more recent one, reducing target movements and patient discomfort.

61 RADIATION PROTECTION AND DOSIMETRY↗
DOE OSTI2018

Dose-volume and radiobiological dependence on the calculation grid size in prostate VMAT planning

This study evaluated the effects of dose-volume and radiobiological dependence on the calculation grid size in prostate volumetric-modulated arc therapy (VMAT) planning. Ten patients with prostate cancer were selected for this retrospective treatment planning study. Prostate VMAT plans were created for the patients using the 6 MV photon beam produced by a Varian TrueBEAM linac with the calculation grid size equal to 1, 2, 2.5, 3, 4, and 5 mm. Dose-volume histograms (DVHs) of targets and organs at risk were generated for different grid sizes. We calculated the radiobiological parameters of the tumor control probability (TCP) of clinical target volume (CTV) and planning target volume (PTV), and the normal tissue complication probability (NTCP) of organs at risk (rectal wall, rectum, bladder wall, bladder, left femur, and right femur). The homogeneity, conformity, and gradient indexes of CTV and PTV were calculated for different grid sizes. The TCP of PTV was found decreasing with a rate of 0.06%/mm when the calculation grid size increased from 1 to 5 mm. On the other hand, both NTCPs of rectal wall and rectum were found decreasing with rates of 0.03%/mm and 0.05%/mm, respectively, with an increase of grid size. The homogeneity index of PTV increased with a rate of 0.57/mm of the calculation grid size, whereas the conformity index of PTV decreased with a rate of 0.0075/mm. The gradient index of PTV was found increasing with a rate equal to 0.05/mm. In prostate VMAT planning, variations of dose-volume and radiobiological parameters with calculation grid size on PTV, rectal wall, and rectum were more significant than those of CTV and other organs at risk such as bladder wall, bladder, and femurs. Results in this study are important in the treatment planning quality assurance when the calculation grid size is varied to compromise a shorter dose computing time.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2017

Boundary Conditions for Diffusion-Mediated Processes within Linear Nanopores: Exact Treatment of Coupling to an Equilibrated External Fluid

In this paper, we consider a variety of diffusion-mediated processes occurring within linear nanopores, but which involve coupling to an equilibrated external fluid through adsorption and desorption. By determining adsorption and desorption rates through a set of tailored simulations, and by exploiting a spatial Markov property of the models, we develop a formulation for performing efficient pore-only simulations of these processes. Coupling to the external fluid is described exactly through appropriate nontrivial boundary conditions at the pore openings. This formalism is applied to analyze the following: (i) tracer counter permeation (TCP) where different labeled particles adsorb into opposite ends of the pore and establish a nonequilibrium steady state; (ii) tracer exchange (TE) with exchange of differently labeled particles within and outside the pore; (iii) catalytic conversion reactions where a reactant in the external fluid adsorbs into the pore and converts to a product which may desorb. The TCP analysis also generates a position-dependent generalized tracer diffusion coefficient, the form of which controls behavior in the TE and catalytic conversion processes. We focus on the regime of single-file diffusion within the pore which produces the strongest correlations and largest deviations from mean-field type behavior. Finally, behavior is quantified precisely via kinetic Monte Carlo simulations but is also captured with appropriate analytic treatments.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2011

Nano-Hydroxyapatite/Fluoridated and Unfluoridated Bioactive Glass Composites: Structural Analysis and Bioactivity Evaluation

Biphasic bioceramic composites containing nano-hydroxyapatite (HAP) and nanosized bioactive glasses have been prepared in the form of pellets and have been examined for the effects of bioglass concentrations and sintering temperature on the structural transformations and bioactivity behavior. Pure stoichiometric nano-HAP was synthesized using sol-gel technique. Two bioglasses synthesized in this work--fluoridated bioglass (Cao-P{sub 2}O{sub 5}-Na{sub 2}O{sub 3}-CaF{sub 2}) and unfluoridated bioglass (Cao-P{sub 2}O{sub 5}-Na{sub 2}O{sub 3}) designated as FBG and UFBG respectively, were added to nano-HAP with concentrations of 5, 10, 12 and 15%. The average particle sizes of synthesized HAP and bioglasses were 23 nm and 35 nm, respectively. The pellets were sintered at four different temperatures i.e. 1000 deg. C, 1150 deg. C, 1250 deg. C and 1350 deg. C. The investigations involved study of structural and bioactivity behavior of green and sintered pellets and their deviations from original materials i.e. HAP, FBG and UFBG, using X-ray diffraction (XRD) and scanning electron microscopy (SEM). The phase composition of the sintered pellets was found to be non-stoichiometric HAP with {alpha}-TCP (tricalcium phosphate) and {beta}-TCP. It was revealed from SEM images that bonding mechanism was mainly solid state sintering for all pellets sintered at 1000 deg. C and 1150 deg. C and also for pellets with lower concentrations of bioglass i.e. 5% and 10% sintered at 1250 deg. C. Partly liquid phase sintering was observed for pellets with higher bioglass concentrations of 12% and 15% sintered at 1250 deg. C and same behaviour was noted for pellets at all concentrations of bioglasses at 1350 deg. C. The sintered density, hardness and compression strength of pellets have been influenced both by the concentration of the bioglasses and sintering temperature. It was observed that the biological HAP layer formation was faster on the green pellets surface than on pure HAP and sintered pellets, showing higher bioactivity in the green pellets.

36 MATERIALS SCIENCE↗
DOE OSTI2012

Dosimetric comparison of volumetric modulated Arc therapy, step-and-shoot, and sliding window IMRT for prostate cancer

This study aims to evaluate treatment plans generated by Step-and-Shoot (SS), Sliding Window (SW) and Volumetric Modulated Arc Therapy (VMAT) in order to assess the differences in dose volume histograms of planning target volume (PTV) and organs at risk (OAR), conformity indices, radiobiological evaluations, and plan quality for prostate cancer cases. Six prostate cancer patients treated in our center were selected for this retrospective study. Treatment plans were generated with Eclipse version 8.9 using 10 MV photon beams. For VMAT, Varian Rapid Arc with 1 or 2 arcs, and for SS and SW IMRT, 7-9 fields were used. Each plan had three PTVs with prescription doses of 81, 59.4, and 45 Gy to prostate, to prostate and lymph nodes, and to pelvis, respectively. Doses to PTV and OAR and the conformal indices (COIN) were compared among three techniques. The equivalent uniform dose (EUD), tumor control probability (TCP) and normal tissue complication probability (NTCP) were calculated and compared. The mean doses to the PTV prostate on average were 83 Gy and the percent differences of mean dose among all techniques were below 0.28. For bladder and rectum, the percent differences of mean dose among all techniques were below 2.2. The COIN did not favour any particular delivery method over the other. The TCP was higher with SS and SW for four patients and higher with VMAT for two patients. The NTCP for the rectum was the lowest with VMAT in five out of the six patients. The results show similar target coverage in general.

61 RADIATION PROTECTION AND DOSIMETRY↗
DOE OSTI2020

Generalized hydrodynamic analysis of transport through a finite open nanopore for two-component single-file systems

Single-file diffusion (SFD) in finite open nanopores is characterized by nonzero spatially varying tracer diffusion coefficients within a generalized hydrodynamic description. This contrasts with infinite SFD systems where tracer diffusivity vanishes. In standard tracer counterpermeation (TCP) analysis, two reservoirs, each containing a different species, are connected to opposite ends of a finite pore. Here, we implement an extended TCP analysis to allow the two reservoirs to contain slightly different mixtures of the two species. Then, determination of diffusion fluxes through the pore allows extraction of diffusion coefficients for near-constant partial concentrations of the two species. This analysis is applied for a lattice-gas model describing two-component SFD through a finite linear pore represented by a one-dimensional array of cells. Two types of particles, A and B, can hop only to adjacent empty cells with generally different rates, $h_A$ and $h_B$. Particles are noninteracting other than exclusion of multiple cell occupancy. Results reveal generalized hydrodynamic tracer diffusion coefficients which adopt small values inversely proportional to pore length in the pore center, but which are strongly enhanced near pore openings.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2017

Tracer counterpermeation analysis of diffusivity in finite-length nanopores with and without single-file dynamics

Here, we perform a tracer counterpermeation (TCP) analysis for a stochastic model of diffusive transport through a narrow linear pore where passing of species within the pore is inhibited or even excluded (single-file diffusion). TCP involves differently labeled but otherwise identical particles from two decoupled infinite reservoirs adsorbing into opposite ends of the pore, and desorbing from either end. In addition to transient behavior, we assess steady-state concentration profiles, spatial correlations, particle number fluctuations, and diffusion fluxes through the pore. From the profiles and fluxes, we determine a generalized tracer diffusion coefficient D tr (x), at various positions x within the pore. D tr (x) has a plateau value in the pore center scaling inversely with the pore length, but it is enhanced near the pore openings. The latter feature reflects the effect of fluctuations in adsorption and desorption, and it is also associated with a nontrivial scaling of the concentration profiles near the pore openings.

36 MATERIALS SCIENCE↗
DOE OSTI2011

Implementing High Performance Remote Method Invocation in CCA

We report our effort in engineering a high performance remote method invocation (RMI) mechanism for the Common Component Architecture (CCA). This mechanism provides a highly efficient and easy-to-use mechanism for distributed computing in CCA, enabling CCA applications to effectively leverage parallel systems to accelerate computations. This work is built on the previous work of Babel RMI. Babel is a high performance language interoperability tool that is used in CCA for scientific application writers to share, reuse, and compose applications from software components written in different programming languages. Babel provides a transparent and flexible RMI framework for distributed computing. However, the existing Babel RMI implementation is built on top of TCP and does not provide the level of performance required to distribute fine-grained tasks. We observed that the main reason the TCP based RMI does not perform well is because it does not utilize the high performance interconnect hardware on a cluster efficiently. We have implemented a high performance RMI protocol, HPCRMI. HPCRMI achieves low latency by building on top of a low-level portable communication library, Aggregated Remote Message Copy Interface (ARMCI), and minimizing communication for each RMI call. Our design allows a RMI operation to be completed by only two RDMA operations. We also aggressively optimize our system to reduce copying. In this paper, we discuss the design and our experimental evaluation of this protocol. Our experimental results show that our protocol can improve RMI performance by an order of magnitude.

99 GENERAL AND MISCELLANEOUS↗
DOE OSTI2022

Throughput Analytics of Cloud Networks

A network of virtual machines at cloud server sites connected over virtual IO connections is a flexible, easily deployable, and cost-effective alternative to a physical network infrastructure with dedicated servers connected over leased fiber lines. We study the throughput performance of such a cloud network by collecting measurements over Google Cloud infrastructure spanning multiple continents. To study its ideal performance and impact of packet losses, we utilize its emulation using dedicated servers and connection hardware emulation devices. We compare the measurements over the cloud network to those over its emulation on a testbed. We examine the throughput profiles of both networks as a function of the round trip time and their utilization-concavity coefficients, estimated using measurements for common TCP versions. The throughput profile's concave-convex shape and its coefficient are critical indicators of the network performance, qualitatively and quantitatively, respectively. The results indicate their overall agreement between the production cloud network and its emulation using dedicated connections, and a near optimal throughput performance of the former except for a few under-performing connections. Also, the number of parallel flows is found to be a dominant factor in optimizing the throughput across various conditions and TCP versions.

Phanekham, Derek↗
DOE OSTI2017

Experiments and Analyses of Data Transfers Over Wide-Area Dedicated Connections

Dedicated wide-area network connections are increasingly employed in high-performance computing and big data scenarios. One might expect the performance and dynamics of data transfers over such connections to be easy to analyze due to the lack of competing traffic. However, non-linear transport dynamics and end-system complexities (e.g., multi-core hosts and distributed filesystems) can in fact make analysis surprisingly challenging. We present extensive measurements of memory-to-memory and disk-to-disk file transfers over 10 Gbps physical and emulated connections with 0–366 ms round trip times (RTTs). For memory-to-memory transfers, profiles of both TCP and UDT throughput as a function of RTT show concave and convex regions; large buffer sizes and more parallel flows lead to wider concave regions, which are highly desirable. TCP and UDT both also display complex throughput dynamics, as indicated by their Poincare maps and Lyapunov exponents. For disk-to-disk transfers, we determine that high throughput can be achieved via a combination of parallel I/O threads, parallel network threads, and direct I/O mode. Our measurements also show that Lustre filesystems can be mounted over long-haul connections using LNet routers, although challenges remain in jointly optimizing file I/O and transport method parameters to achieve peak throughput.

Rao, Nageswara S.↗