diff --git a/CHANGELOG.md b/CHANGELOG.md
index 6cbdb9930..78864eb1c 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -1,9 +1,11 @@
 # CHANGELOG
 
-0.2.0 / Unreleased
+0.2.0 / 2015.03.31
 ==================
 
 - Changelog update, [#9][] [@miketheman][]
+- Fixes `threadstats` unsafe thread operations, [#6][]
+- Add tests to check `statsd` and `threadstats` thread safety [#6][]
 
 0.1.2 / 2015.03.23
 ==================
@@ -16,6 +18,7 @@
 - First release
 
 <!--- The following link definition list is generated by PimpMyChangelog --->
+[#6]: https://github.com/DataDog/datadogpy/issues/6
 [#7]: https://github.com/DataDog/datadogpy/issues/7
 [#9]: https://github.com/DataDog/datadogpy/issues/9
 [@miketheman]: https://github.com/miketheman
diff --git a/README.md b/README.md
index ab616fe4c..4023ebf02 100644
--- a/README.md
+++ b/README.md
@@ -48,7 +48,7 @@ api.Event.create(title=title, text=text, tags=tags)
 from datadog import statsd
 
 statsd.increment('whatever')
-statsd.gauge('foo', 2)
+statsd.gauge('foo', 42)
 
 # Or ThreadStats, an alternative tool to collect and flush metrics, using Datadog REST API
 from datadog import ThreadStats
@@ -57,3 +57,7 @@ stats.start()
 stats.increment('home.page.hits')
 
 ```
+
+Threadsafety
+------------
+`DogStatsd` and `ThreadStats` are threadsafe.
diff --git a/datadog/threadstats/metrics.py b/datadog/threadstats/metrics.py
index dbf5c40ba..330739ab2 100644
--- a/datadog/threadstats/metrics.py
+++ b/datadog/threadstats/metrics.py
@@ -3,6 +3,9 @@
 """
 from collections import defaultdict
 import random
+import itertools
+
+from datadog.util.compat import iternext
 
 
 class Metric(object):
@@ -47,13 +50,14 @@ def __init__(self, name, tags, host):
         self.name = name
         self.tags = tags
         self.host = host
-        self.count = 0
+        self.count = []
 
     def add_point(self, value):
-        self.count += value
+        self.count.append(value)
 
     def flush(self, timestamp):
-        return [(timestamp, self.count, self.name, self.tags, self.host)]
+        count = sum(self.count, 0)
+        return [(timestamp, count, self.name, self.tags, self.host)]
 
 
 class Histogram(Metric):
@@ -67,8 +71,9 @@ def __init__(self, name, tags, host):
         self.host = host
         self.max = float("-inf")
         self.min = float("inf")
-        self.sum = 0
-        self.count = 0
+        self.sum = []
+        self.iter_counter = itertools.count()
+        self.count = iternext(self.iter_counter)
         self.sample_size = 1000
         self.samples = []
         self.percentiles = [0.75, 0.85, 0.95, 0.99]
@@ -76,12 +81,12 @@ def __init__(self, name, tags, host):
     def add_point(self, value):
         self.max = self.max if self.max > value else value
         self.min = self.min if self.min < value else value
-        self.sum += value
+        self.sum.append(value)
         if self.count < self.sample_size:
             self.samples.append(value)
         else:
             self.samples[random.randrange(0, self.sample_size)] = value
-        self.count += 1
+        self.count = iternext(self.iter_counter)
 
     def flush(self, timestamp):
         if not self.count:
@@ -101,7 +106,8 @@ def flush(self, timestamp):
         return metrics
 
     def average(self):
-        return float(self.sum) / self.count
+        sum_metrics = sum(self.sum, 0)
+        return float(sum_metrics) / self.count
 
 
 class Timing(Histogram):
diff --git a/datadog/util/compat.py b/datadog/util/compat.py
index fa617dd65..7aac77d90 100644
--- a/datadog/util/compat.py
+++ b/datadog/util/compat.py
@@ -20,6 +20,10 @@ def is_p3k():
 
     def iteritems(d):
         return iter(d.items())
+
+    def iternext(iter):
+        return next(iter)
+
 else:
     get_input = raw_input
     import ConfigParser as configparser
@@ -29,6 +33,10 @@ def iteritems(d):
     def iteritems(d):
         return d.iteritems()
 
+    def iternext(iter):
+        return iter.next()
+
+
 try:
     from UserDict import IterableUserDict
 except ImportError:
@@ -49,4 +57,3 @@ def iteritems(d):
     import simplejson as json
 except ImportError:
     import json
-
diff --git a/setup.py b/setup.py
index 374f65cc9..d54b31e6d 100644
--- a/setup.py
+++ b/setup.py
@@ -14,7 +14,7 @@
 
 setup(
     name="datadog",
-    version="0.1.2",
+    version="0.2.0",
     install_requires=install_reqs,
     tests_require=["tox"],
     packages=[
diff --git a/tests/performance/test_statsd_thread_safety.py b/tests/performance/test_statsd_thread_safety.py
new file mode 100644
index 000000000..6986698e0
--- /dev/null
+++ b/tests/performance/test_statsd_thread_safety.py
@@ -0,0 +1,59 @@
+import time
+import six
+import threading
+from collections import deque
+from nose import tools as t
+
+from datadog.dogstatsd.base import DogStatsd
+
+
+class FakeSocket(object):
+    """ A fake socket for testing. """
+
+    def __init__(self):
+        self.payloads = deque()
+
+    def send(self, payload):
+        assert type(payload) == six.binary_type
+        self.payloads.append(payload)
+
+    def recv(self):
+        try:
+            return self.payloads
+        except IndexError:
+            return None
+
+    def __repr__(self):
+        return str(self.payloads)
+
+
+class DogstatsdTest(DogStatsd):
+    def send_metrics(self):
+        self.increment('whatever')
+
+
+class TestDogStatsdThreadSafety(object):
+
+    def setUp(self):
+        self.socket = FakeSocket()
+
+    def recv(self):
+        return self.socket.recv()
+
+    def test_send_metrics(self):
+        statsd = DogstatsdTest()
+        statsd.socket = self.socket
+        for _ in range(10000):
+            threading.Thread(target=statsd.send_metrics).start()
+        time.sleep(1)
+        t.assert_equal(10000, len(self.recv()), len(self.recv()))
+
+    def test_send_batch_metrics(self):
+        with DogstatsdTest() as batch:
+            batch.socket = self.socket
+            for _ in range(10000):
+                threading.Thread(target=batch.send_metrics).start()
+        time.sleep(1)
+        payload = map(lambda x: x.split("\n"), self.recv())
+        payload = reduce(lambda prev, ele: prev + ele, payload, [])
+        t.assert_equal(10001, len(payload), len(payload))
diff --git a/tests/performance/test_threadstats_thread_safety.py b/tests/performance/test_threadstats_thread_safety.py
new file mode 100644
index 000000000..6d1e0d0c7
--- /dev/null
+++ b/tests/performance/test_threadstats_thread_safety.py
@@ -0,0 +1,90 @@
+import re
+import time
+import threading
+from nose import tools as t
+
+from datadog import ThreadStats
+
+
+class MemoryReporter(object):
+    """ A reporting class that reports to memory for testing. """
+
+    def __init__(self):
+        self.metrics = []
+        self.events = []
+
+    def flush_metrics(self, metrics):
+        self.metrics += metrics
+
+    def flush_events(self, events):
+        self.events += events
+
+
+class ThreadStatsTest(ThreadStats):
+    def send_metrics_and_event(self, id):
+        # Counter
+        self.increment("counter", timestamp=12345)
+        time.sleep(0.001)  # sleep makes the os continue another thread
+
+        # Gauge
+        self.gauge("gauge_" + str(id), 42)
+        time.sleep(0.001)  # sleep makes the os continue another thread
+
+        # Histogram
+        self.histogram("histogram", id, timestamp=12345)
+        time.sleep(0.001)  # sleep makes the os continue another thread
+
+        # Event
+        self.event("title", "content")
+
+
+class TestThreadStatsThreadSafety(object):
+
+    def test_threadstats_thread_safety(self):
+        stats = ThreadStatsTest()
+        stats.start(roll_up_interval=10, flush_in_thread=False)
+        reporter = stats.reporter = MemoryReporter()
+
+        for i in range(10000):
+            threading.Thread(target=stats.send_metrics_and_event, args=[i]).start()
+        # Wait all threads to finish
+        time.sleep(10)
+
+        # Flush and check
+        stats.flush()
+        metrics = reporter.metrics
+        events = reporter.events
+
+        # Overview
+        t.assert_equal(len(metrics), 10009, len(metrics))
+
+        # Sort metrics
+        counter_metrics = []
+        gauge_metrics = []
+        histogram_metrics = []
+
+        for m in metrics:
+            if re.match("gauge_.*", m['metric']):
+                gauge_metrics.append(m)
+            elif re.match("histogram.*", m['metric']):
+                histogram_metrics.append(m)
+            else:
+                counter_metrics.append(m)
+
+        # Counter
+        t.assert_equal(len(counter_metrics), 1, len(counter_metrics))
+        counter = counter_metrics[0]
+        t.assert_equal(counter['points'][0][1], 10000, counter['points'][0][1])
+
+        # Gauge
+        t.assert_equal(len(gauge_metrics), 10000, len(gauge_metrics))
+
+        # Histogram
+        t.assert_equal(len(histogram_metrics), 8, len(histogram_metrics))
+        count_histogram = filter(lambda x: x['metric'] == "histogram.count", histogram_metrics)[0]
+        t.assert_equal(count_histogram['points'][0][1], 10000, count_histogram['points'][0][1])
+        sum_histogram = filter(lambda x: x['metric'] == "histogram.avg", histogram_metrics)[0]
+        t.assert_equal(sum_histogram['points'][0][1], 4999.5, sum_histogram['points'][0][1])
+
+        # Events
+        t.assert_equal(10000, len(events), len(events))