}
        return totalReceiveTime.get() / (double) receives / 1000.0; // Convert to microseconds
    }

    /**
     * Calculate total throughput in MB/s based on total time
     *
     * @param totalTimeSeconds total elapsed time in seconds
     * @return throughput in MB/s
     */
    public double getThroughputMBps(double totalTimeSeconds) {
        if (totalTimeSeconds <= 0.0) {
            return 0.0;

        long total = requestsSent.get();
        if (total == 0)
            return 0.0;
        return (double) errors.get() / total;
    }

    /**
     * Get throughput in bytes per second
     *
     * @return throughput in bps
     */
    public long getThroughput() {
        long duration = System.currentTimeMillis() - establishedTime;
        if (duration == 0)
            return 0;

        try {
            Thread.sleep(10);
        } catch (InterruptedException e) {
            Thread.currentThread().interrupt();
        }

        long throughput = channelInfo.getThroughput();
        assertTrue(throughput > 0);
    }

    @Test
    void testEquals() {
        ChannelInfo other = new ChannelInfo("test-channel", mockTransport, localInterface, remoteInterface);

This algorithm is specifically optimized for machine generated content.
Write throughput is typically at least 500MB/s per CPU core,
and scales with the number of available CPU cores.
Decompression speed is typically at least 1GB/s.

This means that in cases where raw IO is below these numbers
compression will not only reduce disk usage but also help increase system throughput.
Typically, enabling compression on spinning disk systems

    
    SmbFile file = new SmbFile("smb://server/share/largefile.dat", context);
    
    // Measure single channel throughput
    long singleChannelTime = measureReadTime(file, false);
    
    // Measure multi-channel throughput
    long multiChannelTime = measureReadTime(file, true);
    
    // Multi-channel should be faster

 * sum that is used for purposes such as collecting statistics, not for fine-grained synchronization
 * control. Under low update contention, the two classes have similar characteristics. But under
 * high contention, expected throughput of this class is significantly higher, at the expense of
 * higher space consumption.
 *
 * <p>This class extends {@link Number}, but does not define methods such as {@code

- Performance testing infrastructure

## Success Metrics

1. **Functional Completeness**: All features pass Windows Protocol Test Suite
2. **Performance**: 
   - Multi-channel provides >1.5x throughput improvement
   - RDMA reduces latency by >50%
3. **Reliability**: 
   - Persistent handles survive 99% of network interruptions
   - Witness protocol achieves <5 second failover time

	}, nil
}

// To collect RX stats during "mc support perf net"
// RXSample holds the RX bytes for the duration between
// the last peer to connect and the first peer to disconnect.
// This is to improve the RX throughput accuracy.
type netPerfRX struct {
	RX                uint64    // RX bytes
	lastToConnect     time.Time // time at which last peer to connect to us

	if err := json.NewEncoder(w).Encode(aggregateHealStateResult); err != nil {
		writeErrorResponseJSON(ctx, w, toAdminAPIErr(ctx, err), r.URL)
		return
	}
}

// SitePerfHandler -  measures network throughput between site replicated setups
func (a adminAPIHandlers) SitePerfHandler(w http.ResponseWriter, r *http.Request) {
	ctx := r.Context()

	objectAPI, _ := validateAdminReq(ctx, w, r, policy.HealthInfoAdminAction)

 * writes. The {@link Cache#cleanUp} method of the returned cache will also perform maintenance, but
 * calling it should not be necessary with a high throughput cache. Only caches built with {@link
 * #removalListener removalListener}, {@link #expireAfterWrite expireAfterWrite}, {@link
 * #expireAfterAccess expireAfterAccess}, {@link #weakKeys weakKeys}, {@link #weakValues