In this paper, we propose a new shared memory model: Transactional memory Coherence and Consistency (TCC). TCC provides a model in which atomic transactions are always the basic unit of parallel work, communication, memory coherence, and memory reference consistency. TCC greatly simplifies parallel software by eliminating the need for synchronization using conventional locks and semaphores, along with their complexities. TCC hardware must combine all writes from each transaction region in a program into a single packet and broadcast this packet to the permanent shared memory state atomically as a large block. This simplifies the coherence hardware because it reduces the need for small, low-latency messages and completely eliminates the need for conventional snoopy cache coherence protocols, as multiple speculatively written versions of a cache line may safely coexist within the system. Meanwhile, automatic, hardware-controlled rollback of speculative transactions resolves any correctness violations that may occur when several processors attempt to read and write the same data simultaneously. The cost of this simplified scheme is higher interprocessor bandwidth. To explore the costs and benefits of TCC, we study the characteristics of an optimal transaction-based memory system, and examine how different design parameters could affect the performance of real systems. Across a spectrum of applications, the TCC model itself did not limit available parallelism. Most applications are easily divided into transactions requiring only small write buffers, on the order of 4-8 KB. The broadcast requirements of TCC are high, but are well within the capabilities of CMPs and smallscale SMPs with high-speed interconnects.

No abstract

No abstract

Thread-level speculation (TLS)

Background:RAD001 is an oral inhibitor of mTOR, an intracellular protein kinase downstream of the PI3K/Akt signaling pathway. There is evidence to suggest that Akt promotes breast cancer resistance to cytotoxic treatment and thus targeting this pathway with RAD001 may increase the efficacy of standard chemotherapy. We combined RAD001 in a Phase I dose escalation study with Capecitabine (oral fluorpyrimidine), as treatment of metastatic breast cancer. This study evaluated the oral combination for tolerability and safety. Pharmacokinetics of RAD001 was examined in all dosing cohorts.Methods:This was a single center, Phase I trial. Eligible patients had stage IV breast cancer, have not received endocrine or chemotherapy for 3 weeks prior to initiation and not receiving concurrent endocrine therapy. Patients must not have received prior therapy with capecitabine or an mTOR inhibitor and up to 3 prior chemotherapy regimens were allowable. Pharmacokinetic data are collected weekly for first 2 cycles.All patients received capecitabine (825 mg PO BID) and RAD001 in 3 dose-escalation cohorts: 2.5 mg every other day (Cohort 1), 2.5 mg daily (Cohort 2) and 5 mg daily (cohort 3). The primary endpoint is to determine the DLT of RAD001 + capecitabine, secondary endpoints include tumor response. Treatment continues until disease progression or unacceptable toxicity. Efficacy assessed every 6 weeks for first 4 cycles, then every 9 weeks until end of the 1st year of treatment, then every 12 weeks after.Results: As of April 2009, 12 patients with a median age of 58 years (range 34-75) have been treated at 3 dose levels: 3 in the 1st cohort, 6 in the 2nd cohort and 3 in the 3rd cohort. Toxicities in cohort 1 were generally manageable and expected. They most commonly consisted of nausea, hand-foot syndrome and elevated LFTs. 2 patients in cohort 2 had a grade 3/4 neutropenia and responded to dose reduction. 1 patient had a grade 3/4 elevated lipid panel. Other adverse events (all grade 1/2) included leukopenia, neutropenia, anemia, hypoalbuminemia, elevated LFTs, nausea, hand-foot syndrome, fatigue, sensory neuropathy, and thrombocytopenia. Of the 9 evaluable patients, the clinical benefit rate (CR+PR+SD ≥ 24 weeks) in this heavily pretreated group is 44%. No patient had DLT during cycle 1. 3 patients remain on treatment. Enrollment to the 3rd cohort continues.Conclusion:The all-oral regimen of RAD001 with capecitabine is well tolerated. The maximal tolerated dose (MTD) after 3 dose escalations was not reached. The maximal dose of 5 mg per day of RAD001 was well tolerated. Final toxicity data and preliminary efficacy will be presented at the meeting. Citation Information: Cancer Res 2009;69(24 Suppl):Abstract nr 6112.